Large intracranial metastatic tumors treated by Gamma Knife surgery: outcomes and prognostic factors

Clinical article

Full access

Object

The use of radiosurgery has been well accepted for treating small to medium-size metastatic brain tumors (MBTs). However, its utility in treating large MBTs remains uncertain due to potentially unfavorable effects such as progressive perifocal brain edema and neurological deterioration. In this retrospective study the authors evaluated the local tumor control rate and analyzed possible factors affecting tumor and brain edema response.

Methods

The authors defined a large brain metastasis as one with a measurement of 3 cm or more in at least one of the 3 cardinal planes (coronal, axial, or sagittal). A consecutive series of 109 patients with 119 large intracranial metastatic lesions were treated with Gamma Knife surgery (GKS) between October 2000 and December 2012; the median tumor volume was 16.8 cm3 (range 6.0–74.8 cm3). The pre-GKS Karnofsky Performance Status (KPS) score for these patients ranged from 70 to 100. The most common tumors of origin were non–small cell lung cancers (29.4% of cases in this series). Thirty-six patients (33.0%) had previously undergone a craniotomy (1–3 times) for tumor resection. Forty-three patients (39.4%) underwent whole-brain radiotherapy (WBRT) before GKS. Patients were treated with GKS and followed clinically and radiographically at 2- to 3-month intervals thereafter.

Results

The median duration of imaging follow-up after GKS for patients with large MBTs in this series was 6.3 months. In the first follow-up MRI studies (performed within 3 months after GKS), 77 lesions (64.7%) had regressed, 24 (20.2%) were stable, and 18 (15.1%) were found to have grown. Peritumoral brain edema as defined on T2-weighted MRI sequences had decreased in 79 lesions (66.4%), was stable in 21 (17.6%), but had progressed in 19 (16.0%). In the group of patients who survived longer than 6 months (76 patients with 77 MBTs), 88.3% of the MBTs (68 of 77 lesions) had regressed or remained stable at the most recent imaging follow-up, and 89.6% (69 of 77 lesions) showed regression of perifocal brain edema volume or stable condition. The median duration of survival after GKS was 8.3 months for patients with large MBTs. Patients with small cell lung cancer and no previous WBRT had a significantly higher tumor control rate as well as better brain edema relief. Patients with a single metastasis, better KPS scores, and no previous radiosurgery or WBRT were more likely to decrease corticosteroid use after GKS. On the other hand, higher pre-GKS KPS score was the only factor that showed a statistically significant association with longer survival.

Conclusions

Treating large MBTs using either microsurgery or radiosurgery is a challenge for neurosurgeons. In selected patients with large brain metastases, radiosurgery offered a reasonable local tumor control rate and favorable functional preservation. Exacerbation of underlying edema was rare in this case series. Far more commonly, edema and steroid use were lessened after radiosurgery. Radiosurgery appears to be a reasonable option for some patients with large MBTs.

Abbreviations used in this paper:GKS = Gamma Knife surgery; KPS = Karnofsky Performance Status; MBT = metastatic brain tumor; NSCLC = non–small cell lung cancer; SCLC = small cell lung cancer; WBRT = whole-brain radiotherapy.

Object

The use of radiosurgery has been well accepted for treating small to medium-size metastatic brain tumors (MBTs). However, its utility in treating large MBTs remains uncertain due to potentially unfavorable effects such as progressive perifocal brain edema and neurological deterioration. In this retrospective study the authors evaluated the local tumor control rate and analyzed possible factors affecting tumor and brain edema response.

Methods

The authors defined a large brain metastasis as one with a measurement of 3 cm or more in at least one of the 3 cardinal planes (coronal, axial, or sagittal). A consecutive series of 109 patients with 119 large intracranial metastatic lesions were treated with Gamma Knife surgery (GKS) between October 2000 and December 2012; the median tumor volume was 16.8 cm3 (range 6.0–74.8 cm3). The pre-GKS Karnofsky Performance Status (KPS) score for these patients ranged from 70 to 100. The most common tumors of origin were non–small cell lung cancers (29.4% of cases in this series). Thirty-six patients (33.0%) had previously undergone a craniotomy (1–3 times) for tumor resection. Forty-three patients (39.4%) underwent whole-brain radiotherapy (WBRT) before GKS. Patients were treated with GKS and followed clinically and radiographically at 2- to 3-month intervals thereafter.

Results

The median duration of imaging follow-up after GKS for patients with large MBTs in this series was 6.3 months. In the first follow-up MRI studies (performed within 3 months after GKS), 77 lesions (64.7%) had regressed, 24 (20.2%) were stable, and 18 (15.1%) were found to have grown. Peritumoral brain edema as defined on T2-weighted MRI sequences had decreased in 79 lesions (66.4%), was stable in 21 (17.6%), but had progressed in 19 (16.0%). In the group of patients who survived longer than 6 months (76 patients with 77 MBTs), 88.3% of the MBTs (68 of 77 lesions) had regressed or remained stable at the most recent imaging follow-up, and 89.6% (69 of 77 lesions) showed regression of perifocal brain edema volume or stable condition. The median duration of survival after GKS was 8.3 months for patients with large MBTs. Patients with small cell lung cancer and no previous WBRT had a significantly higher tumor control rate as well as better brain edema relief. Patients with a single metastasis, better KPS scores, and no previous radiosurgery or WBRT were more likely to decrease corticosteroid use after GKS. On the other hand, higher pre-GKS KPS score was the only factor that showed a statistically significant association with longer survival.

Conclusions

Treating large MBTs using either microsurgery or radiosurgery is a challenge for neurosurgeons. In selected patients with large brain metastases, radiosurgery offered a reasonable local tumor control rate and favorable functional preservation. Exacerbation of underlying edema was rare in this case series. Far more commonly, edema and steroid use were lessened after radiosurgery. Radiosurgery appears to be a reasonable option for some patients with large MBTs.

About 174,000 cases (8.3 per 100,000 US population) of metastatic brain tumor (MBT) are diagnosed annually in the United States,19 and 6% of patients with newly diagnosed invasive cancer develop brain metastasis as a result of progression of their original cancer.4 Survival rates for patients with MBTs are often poor. However, with aggressive management of primary and metastatic disease, the overall and progression-free survival rates have improved in recent years. The treatment options for MBTs include resection, whole brain radiotherapy (WBRT), and radiosurgery, such as Gamma Knife surgery (GKS). For patients with large MBTs, resection is usually the treatment of choice. It can rapidly decompress the mass effect of MBTs and relieve the neurological deterioration caused by peritumoral edema.

There is an ongoing debate about the optimal management of MBTs. Stereotactic radiosurgery has recently become more accepted as an alternative treatment because many patients with brain metastases are not optimal candidates for resection and some refuse an open procedure. In the last few decades, satisfactory radiosurgical outcomes in treating small to moderate-size MBTs have been reported after long-term follow-up.15 Radiosurgery is used typically to treat brain metastases that are 3 cm or smaller in maximum diameter. Nevertheless, in the RTOG 9508 study, the use of radiosurgery was permitted for one tumor that was up to 4 cm in diameter.1 To date, there have been few reports discussing the specific outcome of treatment of large MBTs.6,20

For this retrospective report, we reviewed the cases involving MBTs larger than 3 cm in diameter treated with GKS at the University of Virginia. We evaluated the efficacy and safety of single-session GKS for treating these large brain metastases. Several clinical parameters were scrutinized as potential predictors of successful treatment of large MBTs.

Methods

Patient Population

We defined a large brain metastasis as one having a diameter of greater than 3 cm in at least one dimension. From October 2000 to December 2012, 109 patients with 119 large MBTs were treated with GKS at the University of Virginia. Follow-up MRI studies, which were performed within 2–3 months after the GKS treatment, were available for all 109 patients. The study was approved by the institutional review board of the University of Virginia.

The median age of these patients was 60.5 years, and the group included 51 males and 58 females. The median tumor volume was 16.8 cm3 (range 6.0–74.8 cm3). Most patients (68.8%%) had multiple intracranial lesions, and 31.2% of patients had metastasis to other organs. The most common tumor of origin was NSCLC (29.4% of cases), followed by breast cancer (22.9%) and melanoma (21.1%). Of note, those patients included in the current series with SCLC had prior WBRT and demonstrated new or progressive tumors that were targeted with GKS. Prior microsurgical resection had been performed for 36 patients (33.0%), and 43 patients (39.4%) had been treated with WBRT before GKS. To relieve their neurological symptoms, 100 patients (91.7%) were being treated with steroid medication prior to GKS. For patients not receiving steroid therapy prior to GKS, one dose of steroid medication was administered at the time of radiosurgery. Preoperative MR images were carefully reviewed to assess tumor volume, tumor components, and the presence of peritumoral edema on T2-weighted or FLAIR MRI sequences (Table 1).

TABLE 1:

Summary of patient and tumor characteristics for 109 patients with 119 large MBTs*

CharacteristicValue
sex (male:female)51:58
median pt age in yrs (range)60.5 (25.8–88.7)
total no. of tumors >3 cm in diameter119
tumor vol in cm3 (range)16.8 (6.0–74.8)
presence of single/multiple MBTs (in 109 pts)
 single MBT34 (31.2%)
 multiple MBTs75 (68.8%)
tumor location (for 119 MBTs)
 frontal29 (24.4%)
 parietal32 (26.9%)
 temporal16 (13.4%)
 occipital14 11.8%)
 insula1 (0.8%)
 basal ganglia/thalamus6 (5.0%)
 cerebellum21 (17.6%)
median KPS score (range)80 (70–100)
extracranial metastasis (in 109 pts)
 yes34 (31.2%)
 no75 (68.8%)
cancer histology (in 109 pts)
 NSCLC32 (29.4%)
 breast cancer25 (22.9%)
 melanoma23 (21.1%)
 GI tract origin8 (7.3%)
 renal cell8 (7.3%)
 gynecological cancer7 (6.4%)
 SCLC5 (4.6%)
 other1 (0.9%)
median imaging follow-up in mos (range)6.3 (0.1–89.5)
median clinical follow-up in mos (range)7.4 (0.1–92.6)
pre-GKS treatment (in 109 pts)
 steroid therapy100 (91.7%)
 WBRT43 (39.4%)
 radiosurgery5 (4.6%)
 resection36 (33.0%)
 chemotherapy99 (90.8%)
median survival in mos (range)8.3 (1.2–89.6)

GI = gastrointestinal; pt = patient.

Gamma Knife Surgery

Radiosurgery was performed using the Leksell Gamma Unit Model C before 2007, and Perfexion (Elekta AB) after 2007. Treatment planning was done using Gamma-Plan software (Elekta AB). The gross tumor volume was defined using pre- and post-contrast MRI sequences. The planned treatment volume was designed to mirror the gross tumor volume without any appreciable expansion. Multi-isocenter dose planning was used. The radiosurgical margin dose (median 18.0 Gy, range 16.0–26.0 Gy) was prescribed at an isodose level of 30%–70% (median 50%). The median maximum dose to the target tumor volume was 40.0 Gy (range 17.1–66.7 Gy) (Table 2).

TABLE 2:

Radiosurgical data based on treatment of 119 MBTs

ParameterMedianMeanRange
treatment vol (cm3)*15.317.27.0–75.9
margin dose (Gy)18.018.616.0–31.0
isodose level (%)50.046.130–70
max dose (Gy)40.041.017.1–66.7

The treatment volume is for the large brain metastasis, not the entire treatment volume.

Follow-Up Imaging and Clinical Evaluation

After GKS, all patients had regular follow-up by MRI and clinical evaluation at 2- to 3-month intervals. The performance status was scored using the Karnofsky Performance Status (KPS) scale. Changes in tumor volume and characteristics were assessed on follow-up neuroimaging. Tumor response was evaluated on contrast-enhanced T1-weighted images, and edema volume was defined as the peritumoral increased signal detected on T2-weighted images. The tumor and brain edema responses to GKS on MRI studies were classified into 3 categories: 1) decreased, if the area of tumor volume (on contrast-enhanced T1-weighted images) plus the peritumoral area of signal hyperintensity on T2-weighted images had lessened by more than 10% of its original size at the time of GKS; 2) a stable volume, if the area was within 10% of the original size; and 3) increased, if the volume had increased by more than 10%.

Tumor and edema volumes were calculated from the sum of the areas contoured on each slice, multiplied by the slice thickness. Based upon our previous work, the trapezoidal rule formula demonstrates that with accurate delineation on at least 5 slices, calculated volume would have an expected error rate of 10% or less.17 Therefore, this kind of measurement generally has an uncertainty of 10% for tomographic imaging used for radiosurgery of a structural target such as a tumor.17

After GKS, the patients' symptoms, signs, steroid dosage, and any other treatment were documented. Patients were followed up every 3 months afterward.

Statistical Analysis

Analysis was performed with traditional bivariate tests: independent t-test, one-way analysis of variance, and Pearson correlation. Categorical variables, such as sex, were analyzed by the chi-square test. Nevertheless, nonparametric tests (Mann-Whitney U test, Kruskal-Wallis H test, and Spearman's rank correlation) were used when the conditions required by classic tests were not obviously met. Significant variables and interaction expansion covariates (we chose p = 0.15 as the cutoff point) were further assessed in both Cox and logistic multivariable analyses as deemed relevant. A p value of 0.05 or less was considered statistically significant. A Kaplan-Meier plot was used to assess overall survival, and a log-rank test was performed for survival comparison. All statistical analyses in this study were performed using commercially available statistical software (IBM SPSS Statistics 20.0).

Results

Tumor Response and Brain Edema Response After GKS

The median duration of post-GKS imaging followup for patients with large MBTs in this series was 6.3 months. In the first 3 months of follow-up, 77 MBTs showed regression, 24 MBTs were stable, and 18 MBTs were in progression. On the other hand, perifocal edema of 79 MBTs regressed, that of 21 MBTs was stable, and that of the other 19 was in progression. Conclusively, 84.9% of the MBTs had regressed or remained stable, and 84.0% of the MBTs had perifocal brain edema volume regression or were in stable condition (Table 3).

TABLE 3:

Post-GKS outcomes at the first of the repeated imaging studies (within 3 months) and the need of further intervention

VariableGKSWBRT+GKSValue (%)
tumor response (in 119 tumors)
 regression (vol decrease >10%)542377 (64.7)
 stable (vol change <10%)141024 (20.2)
 progression (vol increase >10%)71118 (15.1)
brain edema response (in 119 tumors)
 regression (vol decrease >10%)572279 (66.4)
 stable (vol change <10%)101121 (17.6)
 progression (vol increase >10%)81119 (16.0)
further intervention (in 109 pts)
 craniotomy3912 (11.0)
 WBRT boost15318 (16.5)
 steroid dose increase*7916 (35.6)*

Information on steroid prescription was available in 45 patients; the percentage is calculated based on those cases.

For the patients who survived longer than 6 months (76 patients with 77 MBTs), we further evaluated their tumor response and brain edema at 3-month intervals. In the most recent imaging studies obtained in these patients, 88.3% of the MBTs (68 of 77) had regressed or remained stable in size, and 89.6% of the MBTs (69 of 77) were found to have a reduced or stable volume of perifocal brain edema.

In addition, the patients with previous WBRT had worse tumor response and brain edema response. The detailed clinical data are listed in Table 3. Figure 1 illustrates 2 cases in which there was satisfactory tumor regression 3 months after radiosurgery.

Fig. 1.
Fig. 1.

Representative cases from the current series. A: Axial T1-weighted Gd-enhanced (left) and T2-weighted (right) MR images obtained in a 51-year-old woman with colon cancer and MBTs, including 1 large brain metastasis. Her clinical symptoms included headache, insomnia, and apathy, and she had recently developed mild left paresis. She was treated with upfront GKS. Her KPS score at the time of treatment was 80. Two months after GKS, the tumor volume and edema volume were reduced significantly. The patient survived 14.5 months following GKS. B: Axial T1-weighted Gd-enhanced (left) and T2-weighted (right) MR images obtained in a 60-year-old woman with breast cancer and MBTs, including 1 large brain metastasis. Her clinical symptoms included headache, paresthesias, and an unsteady gait. She had no prior WBRT or resection. Her KPS score was 90 at the time of GKS. Three months after GKS, the tumor volume was stable, and the brain edema had dramatically improved. The patient was still alive 7 months after GKS. M = months.

Neurological Outcomes

The median duration of clinical follow-up after GKS was 7.4 months (Table 1). Most of the patients (91 [83.5%] of 109) demonstrated neurological improvement within 3 months. At the last follow-up examination, 19 (25%) of 76 patients had new neurological symptoms or progression of existing neurological symptoms. Local tumor progression with accompanying neurological decline was treated with a craniotomy and tumor resection in 12 patients (11.0%), WBRT boost in 18 (16.5%), and an increased dosage of steroid medication in 16 (35.6%).

Prognostic Factors Associated With Tumor Control, Brain Edema Response, Decreased Steroid Use, and Overall Survival

In the current study, there was no statistically significant relationship between any of the following factors and tumor control or brain edema control: age at the time of GKS, sex, number of intracranial metastatic lesions, extracranial metastasis, initial tumor volume, KPS score, prescribed radiation dose (including maximum dose and margin dose), previous radiosurgery (for other lesions), previous resection, and previous chemotherapy (Table 4). The absence of previous WBRT and the presence of small cell lung cancer (SCLC) histology were favorable factors for tumor control and improvement in brain edema (p < 0.05 for each).

TABLE 4:

Prognostic factors associated with tumor control, brain edema relief, steroid tapering, and overall survival*

FactorTumor ControlEdema ControlSteroid TaperingOverall Survival
UnivariateMultivariateUnivariateMultivariateUnivariateMultivariateUnivariateMultivariate
age (yrs)0.0890.2150.0520.0540.1610.271
sex0.3270.3160.3110.303
multiple or single brain metastases0.2160.1820.0500.0490.1280.473
extracranial metastasis0.4090.2710.2450.493
tumor vol (cm3)0.8500.9850.5390.268
KPS score0.2130.1520.0330.0120.0240.013
max dose (Gy)0.7720.5060.4370.1360.101
margin dose (Gy)0.4630.2430.4440.954
pre-GKS radiosurgery0.1640.1800.0350.0500.251
pre-GKS WBRT0.0220.0500.0370.0390.0010.0010.529
pre-GKS resection0.3030.3670.1630.855
pre-GKS chemotherapy0.2580.2290.1790.0740.073
tumor histology
 NSCLC0.4200.8030.3760.731
 breast cancer0.5080.5080.8380.583
 melanoma0.5090.2000.2910.241
 GI tract origin0.2160.4680.3880.797
 renal cell0.0670.4190.0710.764
 gynecological cancer0.2160.4680.3390.288
 SCLC0.0040.0020.0040.0030.0170.0320.632
 others0.3280.3280.7050.873

All values are p values. Boldface indicates a statistically significant value.

Tumor control includes regressive and stable tumors.

Brain edema control includes regressive and stable edema.

Presence of only a single MBT, higher pre-GKS KPS score, absence of previous radiosurgery or WBRT, and presence of SCLC histology were all independently associated with an increased likelihood of reduced steroid dosage within 2–3 months following GKS (p < 0.05 for each).

In terms of survival in this cohort of patients with large brain metastases, the only prognostic factor was the pre-GKS KPS score. A better KPS score was significantly related to improved overall survival after GKS (p = 0.024) (Fig. 2).

Fig. 2.
Fig. 2.

Kaplan-Meier plots of overall survival for the cohort of patients with large MBTs. The KPS score at the time of GKS was the only statistically significant prognostic factor for overall survival in this patient population. Number of patients reaching each time milestone is shown along the x-axis. Circles indicate censored patients. MD = margin dose; obs = observational cases; TV = tumor volume.

Complications

Radiation necrosis of brain parenchyma after GKS was not seen in this series. However, one patient with NSCLC had a stable tumor but exacerbation of his brain edema, which we would attribute to GKS. Due to poor overall condition as well as systemic disease progression, the patient did not undergo a craniotomy and tumor resection, but he was treated with a prolonged course of steroid therapy, which helped to lessen the symptoms related to his edema. The patient eventually died slightly more than 6 months after GKS without signs of necrosis on his last brain MRI study.

Discussion

Management of Large Brain Metastases

Although the development of MBTs was formerly considered a terminal event, patients can now survive longer as a result of improved local and systemic therapies.5 Standard treatment for patients with MBTs includes resection, WBRT, single- or multisession radiosurgery, or a combination of these modalities depending on the clinical situation.

Stereotactic radiosurgery works well in treating patients with oligo-metastatic disease and small to moderate-size MBTs. It offers favorable local tumor control and often early volume reduction, depending on the particular type of carcinoma.2,3,11,16 After radiosurgery, approximately 1.2% of patients with small to moderate-size MBTs undergo resection due to radiation necrosis and tumor progression.10

Very few studies have specifically evaluated the outcome of radiosurgical treatment of large MBTs. The main concern was that large MBTs usually cause a mass effect and brain edema, and for this reason their size has historically been considered a relative contraindication for GKS. In this situation, neurosurgeons have generally preferred resection. However, not all patients with large MBTs are willing or able to undergo such surgery. Moreover, many neurological symptoms and signs can be temporarily relieved by steroid therapy while the effects of a definitive treatment are realized.

In our case series, most patients with large MBTs who were treated with stereotactic radiosurgery obtained reasonable results. The majority of patients with large brain metastases exhibited significant tumor volume reduction and brain edema relief simultaneously during the initial 2–3 months following GKS. Worsening local peritumoral edema as a result of tumor progression or radiation-induced changes occurs more frequently in patients with large brain metastases than in those with moderate to small ones. In our experience, worsening edema after GKS has occurred in 16.0% of large MBTs. In such cases, resection or WBRT remain treatment options. However, our experience suggests that many patients with large brain metastases and symptoms that are reasonably well controlled by steroid therapy can be treated with GKS and avoid a resection.

Other studies have reported results following radiosurgery for large brain metastases. At 2 months postradiosurgery, Yang and colleagues20 noted that 74% of patients had improved symptoms, and 54% had tapered or were no longer using steroids. In the same series, the local tumor control rate was 91.4%, with 41% of tumors exhibiting a volume reduction of more than 50% at the first imaging follow-up, and the brain edema control rate was 81.4%. In another study, Han and colleagues7 noted that 85% of patients with large brain metastases showed functional improvement or maintained their independent function status 1–4 months after radiosurgery. These reports and our current study suggest that radiosurgery can be used to achieve local control and reduce peritumoral edema in selected patients with brain metastasis over 3 cm in diameter (Table 5).

TABLE 5:

Previous reports of GKS experiences with large MBTs*

Author & Year (no. of cases)Tumor SizeMedian Pt Age (yrs)Median Marginal Dose (Gy)Median Follow-Up Duration (mos)Tumor Control RateBrain Edema Control RateFavorable FactorsMedian Survival (mos)
Yang et al., 2011 (n = 70)>3 cm6216.08.191.4% in 2 mos81.4% in 2 mossingle metastasis, no previous WBRT, tumor vol <16 cm3 (for tumor reduction); single metastasis, no previous WBRT, breast cancer histology (for peritumor edema vol reduction)8.2
Han et al., 20126 (n = 80)>14 cm35913.8NANANAcontrol of primary disease, marginal dose >11 Gy, tumor vol <26 cm3 (for overall survival)7.9
present study (n = 109)>3 cm6118.06.384.7% in 2–3 mos83.9% in 2–3 mosno previous WBRT, SCLC histology (for tumor control); no previous WBRT, SCLC histology (for peritumor edema vol control); single metastasis, better pre-GKS KPS, no previous RS, no previous WBRT, SCLC histology (for steroid tapering); better pre-GKS KPS (for overall survival)8.3

NA = not available.; RS = radiosurgery.

Tumor control includes regressive and stable tumors.

Brain edema control includes regressive and stable edema.

Prediction of Local Tumor Control for Large Brain Metastases Treated With Radiosurgery

The goal in treating MBTs is not cure of widely metastatic cancer; rather, treatment is undertaken in order to keep the patient from succumbing to intracranial disease. Once the local tumor is controlled, patients will have opportunities to be treated for the primary cancers and other extracranial sites of metastatic disease. From our analysis, lack of prior WBRT and the presence of SCLC histology are the most favorable factors for local tumor control of a large brain metastasis following GKS. The radiosensitivity of SCLC makes tumor control after GKS more likely. Both Yang et al.20 and Han et al.6,7 found that patients without prior WBRT had a higher likelihood of tumor volume stability or reduction than those who had been treated with WBRT. Others factors, including age, sex, numbers of intracranial metastases, extracranial metastasis, and radiation dose, were not related to local tumor control of large brain metastasis after GKS (Table 5).

Patients with large MBTs usually have shorter duration of survival.6,7,20 In the present study, the median survival, based on Kaplan-Meier analysis, was 8.3 months. The only prognostic factor was the pre-GKS KPS score, with those patients who had a better KPS score exhibiting longer survival. In the studies by Han and colleagues,6,7 prolonged survival of patients in these reports was related to the control of primary disease, marginal dose > 11 Gy, and tumor volume < 26 cm2 (Table 5). KPS and control of systemic disease are often related factors and play a significant role in predicting survival—not just in patients with large brain metastasis, but in patients with smaller ones, too.

GKS and Peritumoral Edema Associated With Large Brain Metastases

In our study, tumor volume reduction, and a significant reduction in edema volume occurred in 63.0% of tumors (75 of 119) simultaneously within 3 months of GKS. The extent of edema relief, whether mild or significant, was substantially related to the underlying tumor response following GKS. The patients' neurological symptoms were usually improved and corresponded to the findings of follow-up MRI studies. From the pathophysiological perspective, MBT-induced brain edema is due to vasogenic causes, not cytotoxic ones. Available evidence points to a group of proteinaceous vascular permeability factors18 (for example, VEGF) associated with increased regional capillary permeability in electron micrograph and protein-bound tracer studies.8,13 Once radiosurgery interrupts the vicious cycle of vasogenic edema, symptomatic relief of brain edema can be expected. The current study, however, shows that worsening edema following GKS appears to be more likely with treatment of a large brain metastasis than with treatment of smaller ones. Worsening edema associated with adverse radiosurgical changes and/or tumor progression can be treated with resection, pentoxifylline, vitamin E, or WBRT, along with concomitant steroid therapy.

What Size is “Too Big”?

There is certainly an upper limit at which the risks will exceed the benefits of radiosurgery for brain metastasis. This upper limit is almost certainly going to be defined by a volume rather than a linear dimension, although the neurosurgical literature is filled with linear dimension cutoffs for radiosurgery, such as the traditional 3-cm dogma. The brain metastasis size limit is likely lower for single-session radiosurgery than for multisession radiosurgery. All patients in the current study were treated with single-session GKS. The study was not designed to define the limit, and any upper limit derived from the current data would not be well supported. However, other studies do point to an upper limit. For example, Higuchi et al. treated brain metastases up to 25 cm3 in volume using 3-session GKS.9 Using hypofractionated stereotactic radiotherapy, Jiang et al. treated metastases up to 5.5 cm in diameter.12 Nishizaki et al. used multisession CyberKnife radiosurgery to treat brain metastases up to 53.2 cm3 in volume.14 Future studies will have to better define the upper limit in volume for single- and multisession radiosurgery to a brain metastasis.

Study Limitation

The current study is a retrospective study. Inherent to this study design, there exist patient selection and treatment biases reflective of the treating clinicians and the institution in which we practice. For example, the selection factors in determining the treatment cohort (KPS score > 70) probably meant that the lesions were less likely to be in an eloquent brain location. Also, the decision to employ preradiosurgical WBRT was not always arrived at in a uniform fashion, as this decision was made at times by community physicians. This nonuniformity of WBRT in the preradiosurgical setting for some patients could have introduced a bias into the study.

Conclusions

Although a brain metastasis with a diameter larger than 3 cm has been a relative contraindication for radiosurgery, GKS does appear to offer a reasonable benefit for those patients with large brain metastases who are not suitable candidates for microsurgery or for those in whom a short survival is expected. Our study demonstrated that treatment of large MBTs with radiosurgery resulted in local tumor control and brain edema relief in the majority of patients. At 3 months after GKS, a local tumor control rate of 84.9% and a brain edema control rate of 84.0% were achieved. Radiosurgery may be considered for selected patients with large and clinically challenging MBTs.

Disclosure

Dr. Schlesinger reports receiving support for non–study related clinical or research efforts from Elekta AB.

Author contributions to the study and manuscript preparation include the following. Conception and design: Sheehan, Lee, Yen. Acquisition of data: Sheehan, Lee. Analysis and interpretation of data: Sheehan, Lee, Yen, Xu. Drafting the article: Sheehan, Lee, Yen. Critically revising the article: all authors. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Sheehan. Statistical analysis: Lee, Xu, Schlesinger. Administrative/technical/material support: Sheehan, Yen, Xu. Study supervision: Sheehan.

References

  • 1

    Andrews DWScott CBSperduto PWFlanders AEGaspar LESchell MC: Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 363:166516722004

    • Search Google Scholar
    • Export Citation
  • 2

    Basina BROlson CRoy DKYen CPSchlesinger DNagayama K: Radiation dose and incidence of new metastasis in the anterior temporal lobe structures of radiosurgically treated patients. Clinical article. J Neurosurg 112:1221292010

    • Search Google Scholar
    • Export Citation
  • 3

    Da Silva ANNagayama KSchlesinger DSheehan JP: Early brain tumor metastasis reduction following Gamma Knife surgery. Clinical article. J Neurosurg 110:5475522009

    • Search Google Scholar
    • Export Citation
  • 4

    Davis FGDolecek TAMcCarthy BJVillano JL: Toward determining the lifetime occurrence of metastatic brain tumors estimated from 2007 United States cancer incidence data. Neuro Oncol 14:117111772012

    • Search Google Scholar
    • Export Citation
  • 5

    Gaspar LEScott CMurray KCurran W: Validation of the RTOG recursive partitioning analysis (RPA) classification for brain metastases. Int J Radiat Oncol Biol Phys 47:100110062000

    • Search Google Scholar
    • Export Citation
  • 6

    Han JHKim DGChung HTPaek SHPark CKJung HW: Radiosurgery for large brain metastases. Int J Radiat Oncol Biol Phys 83:1131202012

    • Search Google Scholar
    • Export Citation
  • 7

    Han JHKim DGKim CYChung HTJung HW: Stereotactic radiosurgery for large brain metastases. Prog Neurol Surg 25:2482602012

  • 8

    Hasegawa HUshio YHayakawa TYamada KMogami H: Changes of the blood-brain barrier in experimental metastatic brain tumors. J Neurosurg 59:3043101983

    • Search Google Scholar
    • Export Citation
  • 9

    Higuchi YSerizawa TNagano OMatsuda SOno JSato M: Three-staged stereotactic radiotherapy without whole brain irradiation for large metastatic brain tumors. Int J Radiat Oncol Biol Phys 74:154315482009

    • Search Google Scholar
    • Export Citation
  • 10

    Jagannathan JBourne TDSchlesinger DYen CPShaffrey MELaws ER Jr: Clinical and pathological characteristics of brain metastasis resected after failed radiosurgery. Neurosurgery 66:2082172010

    • Search Google Scholar
    • Export Citation
  • 11

    Jagannathan JYen CPRay DKSchlesinger DOskouian RJPouratian N: Gamma Knife radiosurgery to the surgical cavity following resection of brain metastases. Clinical article. J Neurosurg 111:4314382009

    • Search Google Scholar
    • Export Citation
  • 12

    Jiang XSXiao JPZhang YXu YJLi XPChen XJ: Hypofractionated stereotactic radiotherapy for brain metastases larger than three centimeters. Radiat Oncol 7:362012

    • Search Google Scholar
    • Export Citation
  • 13

    Long DM: Capillary ultrastructure in human metastatic brain tumors. J Neurosurg 51:53581979

  • 14

    Nishizaki TSaito KJimi YHarada NKajiwara KNomura S: The role of cyberknife radiosurgery/radiotherapy for brain metastases of multiple or large-size tumors. Minim Invasive Neurosurg 49:2032092006

    • Search Google Scholar
    • Export Citation
  • 15

    Sheehan JNiranjan AFlickinger JCKondziolka DLunsford LD: The expanding role of neurosurgeons in the management of brain metastases. Surg Neurol 62:32412004

    • Search Google Scholar
    • Export Citation
  • 16

    Sheehan JSchlesinger D: Editorial. Ten brain metastases. J Neurosurg 117:2342362012

  • 17

    Snell JWSheehan JStroila MSteiner L: Assessment of imaging studies used with radiosurgery: a volumetric algorithm and an estimation of its error. Technical note. J Neurosurg 104:1571622006

    • Search Google Scholar
    • Export Citation
  • 18

    Strugar JRothbart DHarrington WCriscuolo GR: Vascular permeability factor in brain metastases: correlation with vasogenic brain edema and tumor angiogenesis. J Neurosurg 81:5605661994

    • Search Google Scholar
    • Export Citation
  • 19

    Walker AERobins MWeinfeld FD: Epidemiology of brain tumors: the national survey of intracranial neoplasms. Neurology 35:2192261985

    • Search Google Scholar
    • Export Citation
  • 20

    Yang HCKano HLunsford LDNiranjan AFlickinger JCKondziolka D: What factors predict the response of larger brain metastases to radiosurgery?. Neurosurgery 68:6826902011

    • Search Google Scholar
    • Export Citation

If the inline PDF is not rendering correctly, you can download the PDF file here.

Article Information

Address correspondence to: Jason Sheehan, M.D., Department of Neurological Surgery, University of Virginia Health System, PO Box 800212, Charlottesville, VA 22908. email: jsheehan@virginia.edu.

Please include this information when citing this paper: published online October 25, 2013; DOI: 10.3171/2013.9.JNS131163.

© AANS, except where prohibited by US copyright law.

Headings

Figures

  • View in gallery

    Representative cases from the current series. A: Axial T1-weighted Gd-enhanced (left) and T2-weighted (right) MR images obtained in a 51-year-old woman with colon cancer and MBTs, including 1 large brain metastasis. Her clinical symptoms included headache, insomnia, and apathy, and she had recently developed mild left paresis. She was treated with upfront GKS. Her KPS score at the time of treatment was 80. Two months after GKS, the tumor volume and edema volume were reduced significantly. The patient survived 14.5 months following GKS. B: Axial T1-weighted Gd-enhanced (left) and T2-weighted (right) MR images obtained in a 60-year-old woman with breast cancer and MBTs, including 1 large brain metastasis. Her clinical symptoms included headache, paresthesias, and an unsteady gait. She had no prior WBRT or resection. Her KPS score was 90 at the time of GKS. Three months after GKS, the tumor volume was stable, and the brain edema had dramatically improved. The patient was still alive 7 months after GKS. M = months.

  • View in gallery

    Kaplan-Meier plots of overall survival for the cohort of patients with large MBTs. The KPS score at the time of GKS was the only statistically significant prognostic factor for overall survival in this patient population. Number of patients reaching each time milestone is shown along the x-axis. Circles indicate censored patients. MD = margin dose; obs = observational cases; TV = tumor volume.

References

  • 1

    Andrews DWScott CBSperduto PWFlanders AEGaspar LESchell MC: Whole brain radiation therapy with or without stereotactic radiosurgery boost for patients with one to three brain metastases: phase III results of the RTOG 9508 randomised trial. Lancet 363:166516722004

    • Search Google Scholar
    • Export Citation
  • 2

    Basina BROlson CRoy DKYen CPSchlesinger DNagayama K: Radiation dose and incidence of new metastasis in the anterior temporal lobe structures of radiosurgically treated patients. Clinical article. J Neurosurg 112:1221292010

    • Search Google Scholar
    • Export Citation
  • 3

    Da Silva ANNagayama KSchlesinger DSheehan JP: Early brain tumor metastasis reduction following Gamma Knife surgery. Clinical article. J Neurosurg 110:5475522009

    • Search Google Scholar
    • Export Citation
  • 4

    Davis FGDolecek TAMcCarthy BJVillano JL: Toward determining the lifetime occurrence of metastatic brain tumors estimated from 2007 United States cancer incidence data. Neuro Oncol 14:117111772012

    • Search Google Scholar
    • Export Citation
  • 5

    Gaspar LEScott CMurray KCurran W: Validation of the RTOG recursive partitioning analysis (RPA) classification for brain metastases. Int J Radiat Oncol Biol Phys 47:100110062000

    • Search Google Scholar
    • Export Citation
  • 6

    Han JHKim DGChung HTPaek SHPark CKJung HW: Radiosurgery for large brain metastases. Int J Radiat Oncol Biol Phys 83:1131202012

    • Search Google Scholar
    • Export Citation
  • 7

    Han JHKim DGKim CYChung HTJung HW: Stereotactic radiosurgery for large brain metastases. Prog Neurol Surg 25:2482602012

  • 8

    Hasegawa HUshio YHayakawa TYamada KMogami H: Changes of the blood-brain barrier in experimental metastatic brain tumors. J Neurosurg 59:3043101983

    • Search Google Scholar
    • Export Citation
  • 9

    Higuchi YSerizawa TNagano OMatsuda SOno JSato M: Three-staged stereotactic radiotherapy without whole brain irradiation for large metastatic brain tumors. Int J Radiat Oncol Biol Phys 74:154315482009

    • Search Google Scholar
    • Export Citation
  • 10

    Jagannathan JBourne TDSchlesinger DYen CPShaffrey MELaws ER Jr: Clinical and pathological characteristics of brain metastasis resected after failed radiosurgery. Neurosurgery 66:2082172010

    • Search Google Scholar
    • Export Citation
  • 11

    Jagannathan JYen CPRay DKSchlesinger DOskouian RJPouratian N: Gamma Knife radiosurgery to the surgical cavity following resection of brain metastases. Clinical article. J Neurosurg 111:4314382009

    • Search Google Scholar
    • Export Citation
  • 12

    Jiang XSXiao JPZhang YXu YJLi XPChen XJ: Hypofractionated stereotactic radiotherapy for brain metastases larger than three centimeters. Radiat Oncol 7:362012

    • Search Google Scholar
    • Export Citation
  • 13

    Long DM: Capillary ultrastructure in human metastatic brain tumors. J Neurosurg 51:53581979

  • 14

    Nishizaki TSaito KJimi YHarada NKajiwara KNomura S: The role of cyberknife radiosurgery/radiotherapy for brain metastases of multiple or large-size tumors. Minim Invasive Neurosurg 49:2032092006

    • Search Google Scholar
    • Export Citation
  • 15

    Sheehan JNiranjan AFlickinger JCKondziolka DLunsford LD: The expanding role of neurosurgeons in the management of brain metastases. Surg Neurol 62:32412004

    • Search Google Scholar
    • Export Citation
  • 16

    Sheehan JSchlesinger D: Editorial. Ten brain metastases. J Neurosurg 117:2342362012

  • 17

    Snell JWSheehan JStroila MSteiner L: Assessment of imaging studies used with radiosurgery: a volumetric algorithm and an estimation of its error. Technical note. J Neurosurg 104:1571622006

    • Search Google Scholar
    • Export Citation
  • 18

    Strugar JRothbart DHarrington WCriscuolo GR: Vascular permeability factor in brain metastases: correlation with vasogenic brain edema and tumor angiogenesis. J Neurosurg 81:5605661994

    • Search Google Scholar
    • Export Citation
  • 19

    Walker AERobins MWeinfeld FD: Epidemiology of brain tumors: the national survey of intracranial neoplasms. Neurology 35:2192261985

    • Search Google Scholar
    • Export Citation
  • 20

    Yang HCKano HLunsford LDNiranjan AFlickinger JCKondziolka D: What factors predict the response of larger brain metastases to radiosurgery?. Neurosurgery 68:6826902011

    • Search Google Scholar
    • Export Citation

TrendMD

Metrics

Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 674 674 34
PDF Downloads 201 201 5
EPUB Downloads 0 0 0

PubMed

Google Scholar