# Brain Metastases: Whole Brain RT, SRS, and Hippocampal Avoidance

## Overview

Brain metastases represent the most common intracranial tumors, affecting between 20% and 40% of patients with cancer. The primary tumors most frequently responsible for brain metastases include lung cancer, which accounts for approximately 50% of cases, followed by breast cancer (15-20%), melanoma (10%), and less commonly renal cell carcinoma and colorectal cancer. Historically, whole brain radiation therapy (WBRT) was the standard treatment for all patients with brain metastases. However, the treatment paradigm has shifted toward stereotactic radiosurgery (SRS) for patients with limited numbers of metastases. Additionally, there is increasing use of hippocampal-avoidant WBRT and strategies employing SRS alone. The choice of treatment depends on multiple factors, including the number and size of metastases, the histology of the primary tumor, the patient’s performance status, the status of extracranial disease, and relevant molecular markers.

## Prognostic Scoring Systems

### Graded Prognostic Assessment (GPA) and Diagnosis-Specific GPA (DS-GPA)

The Graded Prognostic Assessment (GPA) and its updated, diagnosis-specific version (DS-GPA) incorporate molecular markers to better stratify patients according to prognosis. For example, in lung adenocarcinoma, EGFR and ALK mutation status are included; in breast cancer, HER2 and hormone receptor (ER/PR) status are considered; and in melanoma, BRAF mutation status is integrated. Median survival varies widely based on these scores, ranging from as little as 3-4 months in patients with the worst GPA scores to 25-47 months in those with the most favorable prognoses, depending on the primary tumor type. The GPA score helps guide treatment intensity, with aggressive local therapies recommended for patients with favorable scores, while best supportive care may be more appropriate for those with poor prognoses.

## Whole Brain Radiation Therapy (WBRT)

### Technique

The standard WBRT regimen typically involves delivering 30 Gy in 10 fractions, although a shorter course of 20 Gy in 5 fractions is also commonly used. Treatment is delivered using opposed lateral fields that encompass the entire cranial contents. The inferior border of the radiation field is generally set at the base of the skull, around the C1-C2 vertebral junction, ensuring coverage of the cribriform plate anteriorly and the foramen magnum posteriorly. Lens shielding is rarely employed in modern practice because it can compromise adequate coverage of the anterior cranial fossa.

### Indications for WBRT

WBRT remains indicated for patients with multiple brain metastases, historically defined as more than 3-4 lesions, although this threshold is now debated and may extend to patients with more than 5-15 lesions. It is also used in cases of leptomeningeal disease, in patients with poor performance status where the logistics of SRS are impractical, for primary central nervous system lymphoma (a separate clinical entity), and as prophylactic cranial irradiation in small cell lung cancer.

### Declining Role of WBRT

The role of WBRT has diminished due to evidence from trials such as QUARTZ, which compared WBRT plus best supportive care to best supportive care alone in patients with non-small cell lung cancer (NSCLC) brain metastases and poor prognosis (Karnofsky Performance Status < 70). This trial found no significant difference in survival or quality of life between the two groups, indicating that WBRT provides no benefit in this population. Additionally, growing awareness of WBRT-associated neurocognitive toxicity has driven the shift toward SRS-based approaches.

## Stereotactic Radiosurgery (SRS)

### Principles

SRS delivers a high dose of focused radiation to individual brain metastases, either in a single fraction or fractionated over 2 to 5 sessions. It is administered using specialized platforms such as Gamma Knife, CyberKnife, or linear accelerator (linac)-based systems, achieving sub-millimeter targeting accuracy with a steep dose falloff to spare surrounding normal brain tissue.

### Dose Prescriptions (Single Fraction, per RTOG 9005)

Dose prescriptions for single-fraction SRS vary by tumor size. For tumors measuring 2 cm or less, doses of 20-24 Gy are typical. Tumors between 2.1 and 3 cm generally receive 18 Gy, while those between 3.1 and 4 cm receive 15 Gy. For tumors larger than 3 cm or those located near critical structures, fractionated SRS delivering 25-30 Gy over 5 fractions is often considered to balance efficacy and safety.

| Tumor Size | Single-Fraction SRS Dose | Fractionated SRS Alternative |
|---|---|---|
| ≤ 2 cm | 20–24 Gy | — |
| 2.1–3.0 cm | 18 Gy | 25–30 Gy / 5 fx |
| 3.1–4.0 cm | 15 Gy | 25–30 Gy / 5 fx |
| > 4 cm or near critical structures | Generally not single-fraction | 25–30 Gy / 5 fx preferred |
| Postoperative cavity | 12–18 Gy (size-dependent) | 25–30 Gy / 5 fx |

### Evidence for SRS Alone vs. SRS + WBRT

Several randomized trials have compared SRS alone to SRS combined with WBRT. The EORTC 22952-26001 trial studied patients with 1-3 brain metastases and found that while WBRT improved intracranial control, it did not improve overall survival and was associated with increased neurocognitive decline. Similarly, the JROSG 99-1 (Aoyama) trial showed reduced intracranial recurrence with WBRT but no survival benefit. The MD Anderson trial (Chang et al.) comparing SRS alone to SRS plus WBRT was stopped early due to worse neurocognitive outcomes in the WBRT arm. The N0574 trial (Brown et al.) confirmed that SRS alone preserves cognitive function without compromising overall survival compared to SRS plus WBRT. Taken together, these studies support SRS alone as the preferred approach for patients with limited brain metastases (1-4 lesions), provided close MRI surveillance is maintained.

### Expanding the SRS Paradigm: 5-15+ Metastases

Recent retrospective series suggest that SRS can be safely extended to patients with 5 to 15 brain metastases. Importantly, the cumulative intracranial tumor volume appears to be a more relevant factor than the absolute number of metastases. Generally, a total tumor volume less than 15-20 cc is considered manageable with SRS. Prospective data from the JLGK0901 trial (Yamamoto et al.) demonstrated similar overall survival for patients with 2-4 versus 5-10 metastases treated with SRS. However, the upper limit for SRS alone remains a subject of ongoing debate.

## Postoperative Management

### SRS to the Resection Cavity

Surgical resection is indicated for large, symptomatic, or diagnostically uncertain brain metastases. Postoperative SRS directed to the resection cavity significantly reduces local recurrence compared to observation alone. For instance, Mahajan et al. at MD Anderson reported that postoperative SRS decreased the 12-month local recurrence rate from 43% to 28%. Typical dosing ranges from 12-18 Gy in a single fraction to 25-30 Gy over 5 fractions, depending on the size of the resection cavity. While postoperative WBRT also reduces recurrence, it carries a higher risk of neurocognitive decline. The N107C/CEC.3 trial (Brown et al.) compared postoperative SRS to WBRT after resection and found that SRS preserved cognitive function, whereas WBRT provided better intracranial control but no difference in overall survival.

## Hippocampal-Avoidant WBRT

### Rationale

The hippocampal dentate gyrus subgranular zone contains neural stem cells essential for memory formation. Radiation dose to the hippocampus correlates with neurocognitive decline, particularly affecting memory. Brain metastases located within 5 mm of the hippocampus are rare, occurring in approximately 3-5% of cases, making hippocampal avoidance feasible in most patients.

### NRG CC001 Trial

The phase III NRG CC001 trial compared hippocampal-avoidant WBRT (HA-WBRT) plus memantine to standard WBRT plus memantine in patients with brain metastases. HA-WBRT employed conformal avoidance of the hippocampus, maintaining dose constraints of D100% less than 9 Gy and maximum dose under 16 Gy within a 30 Gy in 10 fractions regimen. This approach significantly reduced the risk of cognitive failure at 4 and 6 months without compromising intracranial progression-free survival or overall survival. Consequently, HA-WBRT combined with memantine has become the preferred approach when WBRT is indicated, with the caveat that patients with metastases within 5 mm of the hippocampus were excluded from the trial.

### Memantine

Memantine is an NMDA receptor antagonist that may mitigate glutamate-mediated excitotoxicity to hippocampal neurons during radiation. The RTOG 0614 trial demonstrated that memantine administered during and after WBRT delayed the onset of cognitive decline. Current standard practice includes adding memantine to WBRT at a dose of 20 mg per day, titrated over 1 to 4 weeks and continued for six months.

## Special Considerations by Histology

### Melanoma

Melanoma brain metastases were historically considered radioresistant; however, high single-fraction doses delivered via SRS have proven effective. The advent of immunotherapy agents such as ipilimumab, nivolumab, and pembrolizumab has dramatically improved intracranial response rates. The optimal sequencing of SRS with immunotherapy remains an active area of investigation, but concurrent administration appears safe and may enhance treatment responses.

### HER2-Positive Breast Cancer

Patients with HER2-positive breast cancer have a high lifetime risk (30-50%) of developing brain metastases. The HER2CLIMB trial demonstrated that the combination of tucatinib, trastuzumab, and capecitabine produces intracranial responses. For limited brain metastases, SRS with close surveillance is preferred, and systemic therapy may be used to delay the need for WBRT.

### EGFR-Mutant and ALK-Rearranged NSCLC

In non-small cell lung cancer patients harboring EGFR mutations or ALK rearrangements, CNS-penetrant tyrosine kinase inhibitors (TKIs) such as osimertinib and lorlatinib achieve intracranial response rates of 60-80%. Radiation therapy may be deferred in asymptomatic patients with limited brain metastases who respond to TKIs, reserving SRS for progression on TKI therapy or for symptomatic lesions.

<image>An MRI of the brain showing multiple contrast-enhancing brain metastases (4 lesions of varying sizes in different lobes). Overlaid treatment plan shows SRS dose distributions for each lesion with tight conformality (prescription isodose in red, 50% isodose in blue), demonstrating the ability to treat multiple targets while sparing intervening normal brain tissue. A separate panel shows a single lesion at higher magnification with the steep dose gradient from 100% to 20% over just a few millimeters.</image>

<image>A comparison of standard WBRT versus hippocampal-avoidant WBRT dose distributions on an axial MRI slice at the level of the temporal lobes. The standard WBRT plan shows uniform dose across the entire brain including both hippocampi. The HA-WBRT plan shows IMRT-generated dose reduction over the bilateral hippocampi (contoured in yellow), with the dose constraint annotation showing D100% < 9 Gy and Dmax < 16 Gy. The rest of the brain receives the full 30 Gy prescription.</image>

<image>A clinical decision algorithm flowchart for brain metastases management. Starting with the number of metastases: 1-4 small lesions leads to SRS alone with MRI surveillance; single large/symptomatic lesion leads to surgery followed by cavity SRS; 5-15 lesions with low total volume leads to SRS (with caveats); diffuse/numerous leads to HA-WBRT + memantine (if favorable prognosis) or BSC/short-course WBRT (if poor prognosis per GPA). Molecular markers (EGFR/ALK/HER2/BRAF) branch to consideration of CNS-active systemic therapy.</image>

## Key Clinical Pearls

SRS alone combined with close MRI surveillance has become the standard of care for patients with 1 to 4 brain metastases who have a reasonable prognosis, as adding WBRT does not improve survival but does impair cognitive function. When WBRT is indicated, hippocampal avoidance combined with memantine, as demonstrated in the NRG CC001 trial, should always be employed because there is no justification for delivering standard WBRT if HA-WBRT is technically feasible and the patient has no metastases near the hippocampi. Close MRI surveillance every 2 to 3 months following SRS is essential; the trade-off for avoiding WBRT is the need for frequent imaging and readiness to treat new metastases with salvage SRS. Radiation necrosis occurs in 5-25% of lesions treated with SRS and can be challenging to distinguish from tumor progression on conventional MRI; advanced imaging modalities such as perfusion MRI or PET, along with observation and steroids, are often necessary before deciding on re-treatment. For patients with driver mutations such as EGFR, ALK, HER2, or BRAF, integrating CNS-active systemic therapies into the brain metastasis management plan is crucial, as radiation and systemic therapy are complementary rather than competing modalities. Finally, total intracranial tumor volume is likely a better criterion for selecting patients for SRS than the absolute number of metastases; for example, 10 small metastases totaling 3 cc may be more suitable for SRS than 2 large metastases totaling 30 cc.

## References
- Brown PD et al. "Postoperative stereotactic radiosurgery compared with whole brain radiotherapy for resected metastatic brain disease (NCCTG N107C/CEC.3)." *Lancet Oncol*. 2017;18(8):1049-1060.
- Brown PD et al. "Effect of radiosurgery alone vs radiosurgery with whole brain radiation therapy on cognitive function in patients with 1 to 3 brain metastases: a randomized clinical trial." *JAMA*. 2016;316(4):401-409.
- Gondi V et al. "Hippocampal-avoidant whole-brain radiotherapy plus memantine for patients with brain metastases: NRG Oncology CC001." *J Clin Oncol*. 2022;40(3):232-239.
- Yamamoto M et al. "Stereotactic radiosurgery for patients with multiple brain metastases (JLGK0901)." *Lancet Oncol*. 2014;15(4):387-395.
- Mulvenna P et al. "Dexamethasone and supportive care with or without whole brain radiotherapy in treating patients with non-small cell lung cancer with brain metastases unsuitable for resection or stereotactic radiotherapy (QUARTZ)." *Lancet*. 2016;388(10055):2004-2014.