General Considerations for the Management of Brain Metastases

Brain metastases cause significant morbidity and mortality. Approximately one-quarter of NSCLC patients will develop clinically detectable brain metastases ( ). The outcomes for this population are generally poor; median survival in the absence of treatment is about 1 month and may be extended to 2 months by the use of steroids ( ). Symptoms that are associated with brain metastases may be a result of increased intracranial pressure from cerebral edema, resulting in headaches, nausea, and vomiting. Focal neurological symptoms due to brain metastases include seizures, focal motor and sensory deficits, ataxia, speech problems and cognitive deficits.

Standard treatment options for patients with brain metastases include comfort measures, steroids, anticonvulsants, whole brain radiotherapy (WBRT), stereotactic radiosurgery (SRS), surgical resection, or a combination of these modalities. The optimal treatment for brain metastases is highly dependent on individual patient characteristics, tumor factors, and treatment considerations. The clinician must take these factors into consideration to determine if the goal of treatment is palliation of symptoms or whether long-term disease control with more aggressive treatment modalities might be achievable.

For most patients with multiple brain metastases, WBRT is considered the standard of care. WBRT alone is most commonly used in patients with multiple brain metastases with a limited life expectancy due to poorly controlled extracranial disease or poor performance status. The use of WBRT in NSCLC patients with brain metastases improves quality of life ( ), neurologic function, and results in an improvement in median survival to approximately 3–6 months ( ). Various dose and fractionation schedules for WBRT have been investigated and no altered dose-fractionation schedule has proven to be superior to the commonly prescribed schedules of 20 Gy in five fractions or 30 Gy in 10 fractions with respect to survival or symptom control (neurologic functional status, symptom relief, palliative index or performance status) ( ). Other commonly used fractionation schedules routinely employed in clinical practice are 40 Gy in 20 fractions or 37.5 Gy in 15 fractions.

Management of Patients with Single Brain Metastases

For selected patients with a good performance status and a single brain metastasis, higher rates of local control and survival have been achieved with the use of SRS or surgical resection with WBRT compared to WBRT alone. The first trial to demonstrate a benefit to surgical resection randomized 48 patients with a single brain metastasis and a Karnofsky performance status (KPS) of≥70 to WBRT alone versus surgical resection followed by WBRT ( ). Those patients who received surgery had fewer recurrences at the site of the original metastasis (20 vs 52%, p <0.02), a higher overall median survival (40 vs 15 weeks, p <0.01) and remained functionally independent for longer (38 weeks vs 8 weeks, p <0.005) than those who had WBRT alone.

Stereotactic radiosurgery, a highly precise, typically single-fraction treatment of high dose radiation, has also been shown to improve local control and overall survival when combined with WBRT in patients with a single brain metastasis, as demonstrated by . Three-hundred and thirty-three patients with a KPS of at least 70 and one to three brain metastases from a variety of primary malignancies were randomized to WBRT or WBRT followed by SRS boost to the metastases. Metastases had to be less than 4 cm in diameter for the largest lesion and additional lesions could not exceed 3 cm. The addition of SRS resulted in a local control benefit at 1 year for all patients compared to WBRT alone (82% vs 71%, p =0.01). No survival benefit was observed for patients in the SRS group (6.5 vs 5.7 months) but, on subset analysis, patients with single metastases who received an SRS boost had a significantly better survival (median survival time 6.5 vs 4.9 months, p =0.0393). On multivariate analysis, RTOG RPA Class 1 patients had a survival advantage whether they had a single or multiple tumors. The use of SRS was not associated with any significant additional toxicity.

To summarize, the evidence suggests that the standard of care for patients with a KPS≥70 and a single brain metastasis should be SRS or surgical excision, integrated with adjuvant WBRT. Though these options have never been directly compared to each other in a randomized controlled trial, non-randomized data suggest that both options are considered equally effective. Unfortunately, the only randomized controlled trial designed to address the question failed to accrue enough patients to permit any firm conclusions ( ).

Surgery vs SRS for Single Brain Metastases

The choice of surgery vs SRS depends on several factors including: whether the patient is a surgical candidate; the presence of severe neurologic symptoms; the location, size and resectability of the lesion; and the need for histopathological confirmation. The ASTRO evidence-based guideline on the radiotherapeutic and surgical management of newly diagnosed brain metastases recommends that good prognosis patients with a single brain metastasis less than 3 or 4 cm in maximum dimension and amenable to surgical excision can be considered for either surgery or radiosurgery ( ). In general, surgical resection should be considered in patients who have a surgically accessible lesion, patients with severe neurologic deficits due to mass effect or in patients for whom histopathological confirmation is required ( ). Radiosurgery should be considered in good prognosis patients with single brain metastasis less than 3–4 cm in maximal dimension, in eloquent areas not amenable to safe total resection, or in patients who are unfit for surgery ( ).

Management of Patients with Multiple Brain Metastases

For patients with two to four brain metastases, the randomized data (RCT) have not demonstrated a survival advantage with the addition of SRS to WBRT. This fact is often used as the rationale for treating patients with multiple brain metastases with WBRT alone. Guidelines published by the American Association of Neurosurgeons suggest that the lack of survival benefit observed in the studies examining patients with more than two brain metastases may be due to crossover of study participants and insufficient power in the available RCTs ( ). Furthermore, some non-randomized data have demonstrated a survival advantage for patients with multiple brain metastases treated with SRS. A retrospective study by of 205 patients with four or more intracranial metastases from various primary cancers treated with SRS found a survival benefit compared to historical controls treated with WBRT alone. Total intracranial treatment volume was found to be an important predictor of survival and the authors suggest that this factor be of primary consideration, rather than the total number of metastases.

Though it is unclear whether a survival benefit exists for patients with multiple brain metastases, there is level I evidence that SRS boost results in improved local tumor control, reduced steroid dependence, and improved performance status. randomized 27 patients with two to four brain metastases measuring<2.5 cm and with a KPS≥70 to WBRT alone vs WBRT+SRS. The study was stopped early after an interim analysis revealed substantial improvements in the rate of local failure at 1 year (100% vs 8%) and median time to local failure (6 vs 36 months). No survival benefit was found with the use of SRS boost and WBRT compared to WBRT alone. The RTOG 9508 study ( ), discussed previously, found a statistically significant improvement in KPS and a reduction in steroid use with the addition of SRS to WBRT in patients with one to three brain metastases. This study did not find statistically significant differences in mental status, overall time to intracranial tumor progression, neurologic death rates, or toxicity between the two treatment arms.

Emerging Techniques in the Radiation Management of Brain Metastases

Various radiation therapy approaches (e.g. helical tomotherapy (HT), volumetric arc therapy, and intensity modulated radiation therapy) using image-guided fractionated radiotherapy regimens are being investigated as a non-invasive alternative to SRS in the management of patients with brain metastases. This technique combines intensity modulated fan-beam radiotherapy delivery with megavoltage computed tomography imaging, providing the capability of on-line image-guided conformal radiation delivery. The addition of SRS to WBRT requires separate localization and treatment procedures that add some cost and inconvenience to patients, caregivers and providers. Some SRS systems require invasive immobilization devices, which can add to patient discomfort and patients with multiple lesions can experience long treatment times.

For example, a phase II multi-institutional study ( ) assessing a simultaneous in-field boost HT technique for brain metastases (TOMO-BII) is now open for accrual. Patients with one to three brain metastases are treated with a simultaneous in-field boost technique yielding a total intralesional dose of 60 Gy with a surrounding whole brain dose of 30 Gy. Outcomes will be compared with historical controls treated with SRS and WBRT. The purpose of the trial is to provide patients with a shorter course, dose intense and non-invasive radiotherapy treatment with equivalent or improved CNS control, survival, quality of life and toxicity. Other clinical trials using alternative radiotherapy platforms are also accruing patients ( www.clinicaltrials.gov ).

Surgery for Multiple Brain Metastases

The role of neurosurgical resection for patients with multiple brain metastases remains undefined. The current ASTRO guidelines do not address the utility of surgical resection in patients with more than one brain metastasis. In practice, palliative surgery is considered in patients with multiple brain metastases who have one life-threatening brain lesion that is surgically accessible (i.e. impending herniation, large lesions in critical areas, hemorrhage) to provide rapid decompression or for patients in whom a tissue diagnosis is required. This approach is consistent with the NCCN guidelines on the use of surgery for multiple brain metastases ( http://www.nccn.org/professionals/physician_gls/pdf/cns.pdf .).

Is Whole Brain Radiotherapy Necessary after Surgical Resection or SRS?

Another point of controversy in the brain metastases literature is the need for adjuvant WBRT following surgical resection or SRS. The rationale for the use of adjuvant WBRT is to prevent recurrence in the tumor bed and to control microscopic disease that may be present elsewhere in the brain. The randomized trial by found that the omission of radiotherapy after surgical resection of a single brain metastasis significantly increased the risk of recurrence at the tumor bed and at other sites in the brain. These patients were more likely to die of neurologic causes. randomized 132 patients with one to four brain metastases to SRS and WBRT versus SRS alone. Those patients who did not receive WBRT were more likely to experience intracranial relapse, however, there were no statistically significant differences in the rates of neurologic death.

Thus far, no randomized trial has demonstrated a clear survival benefit to adjuvant WBRT. Given the apparent lack of survival benefit and the concern for the potential long-term neurotoxicity, some investigators advocate withholding WBRT, reserving it for salvage therapy. A recent randomized trial, using a single measure for neurocognitive decline at a single time-point, found that patients with one to three brain metastases treated with SRS and WBRT were found to have a significantly higher risk of deterioration in memory and learning at 4 months, compared to patients who had only SRS ( ). The EORTC 22952-26001 trial compared observation versus WBRT after surgical resection or SRS in 359 patients with one to three brain metastases ( ). The primary endpoint was duration of functional independence. Adjuvant WBRT reduced the intracranial relapse rate and decreased the rate of neurologic death. No significant difference in the duration of functional independence was observed, however, the rate of neurologic death due to intracranial failure was significantly lower in patients who received WBRT. In summary, both of these studies suggest that WBRT may be withheld in patients with one to three brain metastases treated with SRS if frequent serial brain imaging is performed to detect early failures requiring salvage treatment.

In contrast, it is well established that patients who do not receive WBRT are at higher risk for recurrence in the tumor bed and elsewhere in the brain. Whether the morbidity of recurrence is more severe than the morbidity related to WBRT remains unclear, since data on quality of life outcomes and neurocognitive function in patients who do not receive WBRT are limited. In a retrospective study of 36 patients treated with SRS and observation, found that 47% had a recurrence of brain metastases and 71% of these patients were symptomatic. Of those patients with recurrent tumors, 59% had neurologic deficits that did not improve after salvage radiotherapy. These findings suggest that uncontrolled brain tumors can cause a substantial decrease in mental performance that far outweighs any decrement seen with cranial irradiation ( ) and that recurrent brain metastases may be less amenable to successful treatment. Therefore, ASTRO guidelines recommend that treatment options for patients with one to three brain metastases and stable systemic disease include any of the following: SRS or surgical resection alone with serial imaging for follow up; SRS with WBRT; or WBRT alone. For patients with a single brain metastasis treated with surgical resection, it is strongly recommended that adjuvant WBRT be considered, as WBRT reduces the risk of tumor bed recurrence ( ).

Which Patients Benefit from Aggressive Management of Brain Metastases?

Predicting which patients with brain metastases may benefit from aggressive treatment can be very challenging for the clinician. Several major factors have been identified in the literature that influence prognosis and predict for therapeutic benefit. The RTOG developed a recursive partitioning model based on 1960 patients with brain metastases treated with WBRT alone, WBRT and radiosensitizers or WBRT and radiosurgery and divided these patients into three prognostic groups based on pre-treatment prognostic variables: age, performance status, primary tumor status, presence/absence of extracranial systemic metastases ( ). Patients with the highest median survival (7.1 months) were<65 years old, had a KPS≥70, with a controlled primary and no extracranial disease. The poorest median survival of 2.3 months was observed in Class III patients (KPS<70). Class II (all others patient combinations) had a median survival of 4.2 months.

The number of brain metastases is another important prognostic factor. The prognostic significance of the number of brain metastases was demonstrated in the RTOG 9508 randomized controlled trial that compared WBRT and SRS versus WBRT alone in patients with one to three brain metastases ( ). They found a survival benefit in only those patients with a single brain metastasis. No survival benefit was observed in those patients who had two or three brain metastases.

A recent systematic review of brain metastases prognostic indices for a variety of primary cancers found that nine prognostic systems have been published in the medical literature ( ). Four of the nine prognostic systems include control of the primary malignancy as an important prognostic factor. Other factors commonly used in these prognostic indices are performance status, age, number of brain metastases, control of the primary malignancy and presence of extracranial metastases.

Prognostic Factors Specific to NSCLC Patients with Brain Metastases

The RTOG recursive partitioning analysis was refined by to include the primary disease site as a parameter in 4259 patients treated with surgery, WBRT, SRS or various treatment combinations to define a disease-specific graded prognostic assessment (DS-GPA) score. For patients with brain metastases from NSCLC, age, KPS, presence of extracranial metastases, and number of brain metastases were identified as prognostic factors. The median overall survival for all NSCLC patients with brain metastases was 7 months. A median survival of 14.78 months was achieved in patients with the highest DS-GPA score (≤50 years old, KPS 90–100, absence of extracranial metastases and a single brain metastasis). In contrast, those with the lowest DS-GPA score (>60 years, KPS<70, presence of ECM, more than three brain metastases) had a median survival of only 3.02 months.