Does size matter? Investigating the optimal planning target volume margin for postoperative stereotactic radiosurgery to resected brain metastases

View More View Less
  • 1 Departments of Radiation Oncology and
  • 5 Neurosurgery, and
  • 4 Biostatistics and Bioinformatics Shared Resource,
  • 2 Winship Cancer Institute, Emory University, Atlanta, Georgia;
  • 3 Department of Radiation Oncology, Rush University Medical Center, Chicago, Illinois; and
  • 6 Department of Therapeutic Radiology, Yale School of Medicine, New Haven, Connecticut
Full access

OBJECTIVE

The optimal margin size in postoperative stereotactic radiosurgery (SRS) for brain metastases is unknown. Herein, the authors investigated the effect of SRS planning target volume (PTV) margin on local recurrence and symptomatic radiation necrosis postoperatively.

METHODS

Records of patients who received postoperative LINAC-based SRS for brain metastases between 2006 and 2016 were reviewed and stratified based on PTV margin size (1.0 or > 1.0 mm). Patients were treated using frameless and framed SRS techniques, and both single-fraction and hypofractionated dosing were used based on lesion size. Kaplan-Meier and cumulative incidence models were used to estimate survival and intracranial outcomes, respectively. Multivariate analyses were also performed.

RESULTS

A total of 133 patients with 139 cavities were identified; 36 patients (27.1%) and 35 lesions (25.2%) were in the 1.0-mm group, and 97 patients (72.9%) and 104 lesions (74.8%) were in the > 1.0–mm group. Patient characteristics were balanced, except the 1.0-mm cohort had a better Eastern Cooperative Group Performance Status (grade 0: 36.1% vs 19.6%), higher mean number of brain metastases (1.75 vs 1.31), lower prescription isodose line (80% vs 95%), and lower median single fraction–equivalent dose (15.0 vs 17.5 Gy) (all p < 0.05). The median survival and follow-up for all patients were 15.6 months and 17.7 months, respectively. No significant difference in local recurrence was noted between the cohorts. An increased 1-year rate of symptomatic radionecrosis was seen in the larger margin group (20.9% vs 6.0%, p = 0.028). On multivariate analyses, margin size > 1.0 mm was associated with an increased risk for symptomatic radionecrosis (HR 3.07, 95% CI 1.13–8.34; p = 0.028), while multifraction SRS emerged as a protective factor for symptomatic radionecrosis (HR 0.13, 95% CI 0.02–0.76; p = 0.023).

CONCLUSIONS

Expanding the PTV margin beyond 1.0 mm is not associated with improved local recurrence but appears to increase the risk of symptomatic radionecrosis after postoperative SRS.

ABBREVIATIONS ECOG = Eastern Cooperative Group; GTR = gross-total resection; GTV = gross tumor volume; IDL = isodose line; LINAC = linear accelerator; MVA = multivariate analysis; OS = overall survival; PTV = planning target volume; SFED = single fraction–equivalent dose; SRS = stereotactic radiosurgery; STR = subtotal resection; WBRT = whole-brain radiation therapy.

OBJECTIVE

The optimal margin size in postoperative stereotactic radiosurgery (SRS) for brain metastases is unknown. Herein, the authors investigated the effect of SRS planning target volume (PTV) margin on local recurrence and symptomatic radiation necrosis postoperatively.

METHODS

Records of patients who received postoperative LINAC-based SRS for brain metastases between 2006 and 2016 were reviewed and stratified based on PTV margin size (1.0 or > 1.0 mm). Patients were treated using frameless and framed SRS techniques, and both single-fraction and hypofractionated dosing were used based on lesion size. Kaplan-Meier and cumulative incidence models were used to estimate survival and intracranial outcomes, respectively. Multivariate analyses were also performed.

RESULTS

A total of 133 patients with 139 cavities were identified; 36 patients (27.1%) and 35 lesions (25.2%) were in the 1.0-mm group, and 97 patients (72.9%) and 104 lesions (74.8%) were in the > 1.0–mm group. Patient characteristics were balanced, except the 1.0-mm cohort had a better Eastern Cooperative Group Performance Status (grade 0: 36.1% vs 19.6%), higher mean number of brain metastases (1.75 vs 1.31), lower prescription isodose line (80% vs 95%), and lower median single fraction–equivalent dose (15.0 vs 17.5 Gy) (all p < 0.05). The median survival and follow-up for all patients were 15.6 months and 17.7 months, respectively. No significant difference in local recurrence was noted between the cohorts. An increased 1-year rate of symptomatic radionecrosis was seen in the larger margin group (20.9% vs 6.0%, p = 0.028). On multivariate analyses, margin size > 1.0 mm was associated with an increased risk for symptomatic radionecrosis (HR 3.07, 95% CI 1.13–8.34; p = 0.028), while multifraction SRS emerged as a protective factor for symptomatic radionecrosis (HR 0.13, 95% CI 0.02–0.76; p = 0.023).

CONCLUSIONS

Expanding the PTV margin beyond 1.0 mm is not associated with improved local recurrence but appears to increase the risk of symptomatic radionecrosis after postoperative SRS.

ABBREVIATIONS ECOG = Eastern Cooperative Group; GTR = gross-total resection; GTV = gross tumor volume; IDL = isodose line; LINAC = linear accelerator; MVA = multivariate analysis; OS = overall survival; PTV = planning target volume; SFED = single fraction–equivalent dose; SRS = stereotactic radiosurgery; STR = subtotal resection; WBRT = whole-brain radiation therapy.

Brain metastases represent the most common intracranial tumors.8,23 Surgical intervention is often recommended for patients with large, symptomatic lesions or for patients who require a pathologic diagnosis. Postoperative whole-brain radiation therapy (WBRT) has been shown to improve local control and distant brain failure, but not overall survival (OS).13,20 Given concerns of neurocognitive sequelae following WBRT,4,6 postoperative stereotactic radiosurgery (SRS) followed by close surveillance has emerged as an alternative treatment paradigm.3,24

For intact brain metastases, a planning target volume (PTV) margin can be added to account for setup uncertainties related to patient immobilization, choice of image-guided radiotherapy, and delivery approach (e.g., linear accelerator [LINAC]–based SRS vs Gamma Knife radiosurgery).12 In addition to these factors, resection cavities are associated with unique challenges that make target delineation more difficult: irregular size, changes in shape over time, and artifacts due to blood products.1,21 These factors lead to higher rates of interobserver bias. While local control rates can be high (75%–80%), Choi et al. retrospectively demonstrated that adding no margin for SRS to resection cavities is associated with worse local control when compared with a 2.0-mm margin.7 As such, multiple institutions have implemented an additional margin for resection cavities treated with postoperative SRS alone.16,28

Adding a margin to ensure adequate target coverage, however, is not without consequence. Larger margins result in an increased dose delivered to adjacent normal brain tissue. Since the volume of brain receiving 10 and 12 Gy correlates with intracranial necrosis for unresected lesions,2,17 increasing the PTV margin could result in a higher risk of radiation necrosis. Choi et al., however, demonstrated no toxicity differences between a PTV margin of 2.0 mm and no PTV margin.7 It therefore remains to be seen how smaller margins (between 0 and 2.0 mm) compare with regard to local control and radiation necrosis. Highlighting this variability, 2 recent prospective studies used varying PTV margins for resection cavities treated with postoperative SRS: the MD Anderson group using a 1.0-mm PTV margin15 and the Alliance group (study number NC107c) using a 2.0-mm PTV margin.5

With limited research to date dedicated to investigating the safety and efficacy of different PTV margin size, we retrospectively investigated the treatment outcomes of a margin size of 1.0 mm versus > 1.0 mm for patients who underwent resection of their brain metastases and received postoperative SRS.

Methods

Patients

After obtaining IRB approval, we retrospectively reviewed records of patients who were 18 years or older and underwent craniotomy for brain metastases, followed by postoperative SRS from 2006 to 2016. The IRB determined that individual patient consent was not required for this chart review. Patients were ineligible for this study if they had received prior WBRT or had a diagnosis of a radiosensitive tumor (e.g., small cell lung cancer, lymphoma, and germ cell).

All patients’ charts were reviewed for the following characteristics: age at diagnosis, sex, primary tumor histology, preoperative tumor volume on MRI, number of brain metastases, time to surgery, extent of resection classified as gross-total resection (GTR) or subtotal resection (STR), Karnofsky Performance Scale score, Eastern Cooperative Group (ECOG) Performance Status, graded prognostic assessment, and recursive partitioning analysis. The following radiation treatment variables were collected: number of fractions, single fraction–equivalent dose (SFED), technique (framed vs frameless), prescription isodose line (IDL), gross tumor volume (GTV, in cm3), PTV size (cm3), PTV margin (mm), and conformity index. The conformity index was defined as the ratio of the treatment volume and PTV.

Patients were stratified based on PTV margin size. Treated resection cavities were divided into 2 groups: one group with a margin size of 1.0 mm and the other with a margin size > 1.0 mm. For lesions in the > 1.0–mm group, the PTV margin ranged from 1.5 to 3.0 mm, based on physician preference.

Radiation Treatment

Approximately 2–4 weeks after craniotomy, patients underwent CT simulation with a thermoplastic head mask for immobilization (frameless technique) or a rigid head frame (framed technique). A high-resolution CT simulation brain scan was obtained using a SOMATOM Definition AS CT scanner (Siemens Healthineers) with a slice thickness of 0.6 mm. This CT data set was then coregistered with the diagnostic contrast-enhanced MR image (T1-weighted, magnetization-prepared rapid gradient-echo sequences) with a slice thickness of 1.0 mm. The diagnostic MR image was obtained within 1–2 weeks of CT simulation and was used for treatment planning. After accurate rigid image registration, the target volume was contoured on the CT scan with the aid of the MR images.

The GTV was defined as the postoperative resection cavity and area of residual disease in cases of STR; the clinical target volume was identical to the GTV. The clinical target volume was then uniformly expanded to form the PTV, with the margin size chosen by the treating physician. The radiation dose and fractionation were also left to the judgment of the treating physician. Historically, our institution has used single-fraction SRS, with the dose based on results from the Radiation Therapy Oncology Group (RTOG) 90–05,27 but for larger lesions, our clinical practice is to use multifraction radiosurgery to limit toxicity.9,14 To account for differences in fractionation, SFED was calculated for each patient using an α/β ratio of 10.

Patients were exclusively treated with LINAC-based SRS within 1 week of CT simulation. All patients were treated using 6-MV photons. Varian Novalis TX (Varian Medical Systems, Inc.) was used to deliver the radiosurgery treatments with a high-definition multileaf collimator thickness of 2.5 mm. If the patient was treated using a rigid, stereotactic head frame, this was placed at the time of CT simulation by a neurosurgeon, followed by treatment delivery a few hours later.

Follow-Up

Post-SRS follow-up consisted of clinical examination and brain MRI 4–6 weeks later, with subsequent follow-up every 3 months thereafter, unless clinically indicated otherwise. Local recurrence was defined as recurrence seen within the prior 80% IDL of the resection cavity on follow-up MRI. Regional recurrence was defined as the presence of new enhancing lesions distinctly outside the irradiated field. Leptomeningeal disease was defined as new, abnormal leptomeningeal enhancement outside the prior treated region.

Radiographic radiation necrosis was defined as the development of a contrast-enhancing mass within prior radiation treatment fields.22 Cases were discussed at a multidisciplinary neurooncology tumor board, consisting of neurosurgeons, radiation oncologists, neuroradiologists, medical oncologists, and pathologists, to develop a consensus decision differentiating radiation necrosis versus local recurrence. Additional imaging studies (MR perfusion, MR spectroscopy, and brain PET scans) were obtained when a consensus was not reached. Symptomatic radiation necrosis was defined as development of radiation necrosis requiring treatment; steroids were initially used, with resection reserved for nonresponders. In cases of surgery, patients were deemed to have radiation necrosis if there was no evidence of residual disease; otherwise, they were considered to have local recurrence.

Statistical Analysis

Statistical analysis was divided into lesion-level analysis and patient-level analysis. Local recurrence was evaluated at the lesion level, while radiation necrosis and symptomatic radiation necrosis were evaluated at the patient level. Outcome events (local recurrence, regional recurrence, and symptomatic radiation necrosis) were calculated as time from SRS to date of event. OS was defined as death due to any cause, and patients were censored at time of last follow-up.

Fine and Gray’s method was used for multivariate analysis (MVA) of competing risk end points. The Kaplan-Meier method was used to estimate OS, and the log-rank test was used to assess for differences between treatment groups. The Cox proportional hazards model was used to perform MVA to generate hazard ratios with associated 95% confidence intervals. Univariate and multivariate analyses were performed using collected clinical and dosimetric factors to determine their impact on local recurrence, regional recurrence, OS, and symptomatic radiation necrosis. Variables found to be significant on univariate analysis, using a backward selection with an α-level of removal of 0.20, were then entered into MVA. All statistical analyses were performed using SAS (version 9.4, SAS Institute Inc.).

Results

Patient and Lesion Characteristics

Between 2006 and 2016, a total of 133 consecutive patients with 139 resected brain metastases received postoperative SRS at our institution. Table 1 details the patient- and lesion-level characteristics.

TABLE 1.

Baseline patient and lesion characteristics between the 1.0-mm and >1.0–mm margin cohorts

Group
Covariate1.0-mm Margin>1.0–mm Marginp Value
No. of patients/lesions36/3597/104
Age, yrs0.103
 ≤6529 (80.6)64 (66.0)
 >657 (19.4)33 (34.0)
Sex0.261
 Male18 (50.0)38 (39.2)
 Female18 (50.0)59 (60.8)
Primary site0.288
 Lung12 (33.3)47 (48.5)
 Breast7 (19.4)13 (13.4)
 Melanoma10 (27.8)16 (16.5)
 RCC, GI, or other7 (19.4)21 (21.7)
Extracranial disease0.551
 Yes14 (38.9)32 (33.3)
 No22 (61.1)64 (66.7)
ECOG grade0.032
 013 (36.1)19 (19.6)
 121 (58.3)57 (58.8)
 >12 (5.6)21 (21.7)
RPA class0.712
 110 (27.8)33 (34.0)
 225 (69.4)60 (61.9)
 31 (2.8)4 (4.1)
GPA class0.712
 0–1.02 (5.7)3 (2.9)
 1.5–2.517 (48.6)70 (67.3)
 3.06 (17.1)20 (19.2)
 3.5–4.010 (28.6)11 (10.6)
Technique0.003
 Framed33 (91.7)64 (66.0)
 Frameless3 (8.3)33 (34.0)
Time from surgery (days)
 Mean37.838.10.915
 Median3135
No. of brain metastases
 Mean1.751.310.011
 Median1.01.0
No. of fractions
 Mean1.61.70.543
 Median11
 Range3–53–5
GTV size, cm3
 Mean12.614.30.378
 Median10.612.2
PTV size, cm3
 Mean17.722.90.094
 Median14.720.3
PTV margin, mm
 Mean1.01.9<0.001
 Median1.02.0
SFED, Gy
 Mean16.016.90.011
 Median15.017.5
Conformity index
 Mean1.61.50.156
 Median1.51.5
Prescription IDL, %
 Mean88920.003
 Median8095

GI = gastrointestinal; GPA = graded prognostic assessment; RCC = renal cell carcinoma; RPA = recursive partitioning analysis.

Values are presented as the number of patients (%) unless stated otherwise. Boldface type indicates statistical significance.

There were 36 (27.1%) patients in the 1.0-mm margin group and 97 (72.9%) patients in the > 1.0–mm margin group (mean margin size 1.9 mm, median margin size 2.0 mm). Patient characteristics were well balanced, except the 1.0-mm cohort had better ECOG Performance Status (ECOG grade 0: 36.1% vs 19.6%, p = 0.032) and more patients were treated with the framed technique (91.7% vs 66.0%, p = 0.003). The most common histology was non–small cell lung cancer (33.3% and 48.5% in the 1.0-mm and > 1.0–mm groups, respectively). There was no statistically significant difference between the tumor types between the groups.

For lesion-level characteristics, there were 35 (25.2%) lesions within the 1.0-mm margin group and 104 (74.8%) lesions within the > 1.0–mm margin group. The > 1.0–mm margin cohort had a higher median prescription IDL (95% vs 80%, p = 0.003), higher median SFED (17.47 vs 15 Gy, p = 0.042), and median PTV margin size (2.0 vs 1.0, p < 0.001). For multifraction SRS, the number of fractions ranged from 3 to 5. The dose per fraction ranged from 5 to 7 Gy. The median conformity index and cavity size were not statistically different between the 2 groups: 1.5 conformity index in each group and 10.6 vs 12.2 cm3 in the 1.0-mm and > 1.0–mm groups, respectively (p = 0.378).

Locoregional Recurrence

Figure 1 illustrates the cumulative incidence of local recurrence stratified by margin size. The 1-year local recurrence rates were 15.2% and 14.3% for the 1.0-mm and > 1.0–mm groups, respectively (p = 0.91). Significant predictors for local recurrence on univariate analysis were ECOG score of 1 (HR 0.26, 95% CI 0.10–0.64; p = 0.004), GTV > 15 cm3 (HR 4.01, 95% CI 1.67–9.64; p = 0.002), and PTV > 14 cm3 (HR 4.90, 95% CI 1.12–21.41; p = 0.035).

FIG. 1.
FIG. 1.

Cumulative incidence of local recurrence stratified by margin group. Figure is available in color online only.

Table 2 shows the MVA for local recurrence. Margin size and SFED were not statistically significant for local recurrence (HR 1.09 [95% CI 0.42–2.85], p = 0.862 and HR 1.14 [95% CI 0.75–1.74], p = 0.540; respectively). GTV > 15 cm3 (HR 4.23, 95% CI 1.02–17.58; p = 0.047) persisted as a statistically significant variable that predicted for higher rates of local recurrence.

TABLE 2.

MVA for local recurrence

Local Recurrence (mos)
CovariateHR (95% CI)HR p ValueType 3 p Value
Margin size, mm0.862
 11.09 (0.42–2.85)0.862
 >1Ref
GTV, cm30.047
 >154.23 (1.02–17.58)0.047
 ≤15Ref
Extent of resection0.942
 STR1.04 (0.33–3.28)0.942
 GTRRef
No. of fractions0.743
 >10.77 (0.16–3.63)0.743
 1Ref
Conformity index0.55 (0.08–3.75)0.5420.542
SFED1.14 (0.75–1.74)0.5400.540
Age at treatment, yrs0.073
 ≤654.06 (0.88–18.79)0.073
 >65Ref
No. of brain metastases0.140
 12.36 (0.75–7.38)0.140
 >1Ref

Boldface type indicates statistical significance. Number of observations in the original data set = 141. Number of observations used = 117. Backward selection with an α-level of removal of 0.2 was used. The following variables were removed from the model: ECOG, graded prognostic assessment class, time from surgery, sex, PTV size, and presence of extracranial disease.

The median time to regional recurrence was 7 months. There was no statistically significant difference for time to development of regional recurrence between the groups (p = 0.572). As depicted in Supplementary Fig. 2, at 1 year, the probability of regional recurrence in the 1.0-mm margin group was 41.5% versus 49.4% for the > 1.0–mm margin group (p = 0.3603). At 1 year, 17 of 133 patients (actuarial rate of 12.8%) developed recurrence within the leptomeningeal compartment.

Symptomatic Radiation Necrosis

Overall, 28 of 133 patients (21.1%) developed symptomatic radiation necrosis, of whom 24 patients were in the > 1–mm group. Using patient-level analysis, at 1 year, 16.7% of patients developed symptomatic radiation necrosis. These patients were initially treated with oral glucocorticoids. Due to refractory symptoms, 3 patients who developed symptomatic radiation necrosis required surgical management whereby pathology confirmed the diagnosis. Of note, these 3 patients all were treated with a margin size > 1.0 mm (median 1.5 mm). One- and 2-year symptomatic radiation necrosis rates (Fig. 2) were 6.0% and 9.1%, and 20.9% and 26.6% in the 1.0-mm and > 1.0–mm margin cohorts, respectively (p = 0.028). Table 3 shows MVA for symptomatic radionecrosis. MVA confirmed that margin size > 1.0 mm (HR 3.07, 95% CI 1.13–8.34; p = 0.028) was a statistically significant predictor for increased rates of symptomatic radiation necrosis. Furthermore, number of fractions > 1 also emerged as a protective factor for symptomatic radiation necrosis (HR 0.13, 95% CI 0.02–0.76; p = 0.023) on MVA. SRS technique (framed vs frameless) and SFED were not found to be statistically significant factors for symptomatic radiation necrosis (p = 0.145).

FIG. 2.
FIG. 2.

Cumulative incidence of symptomatic radiation necrosis stratified by margin group. Figure is available in color online only.

TABLE 3.

MVA for symptomatic radiation necrosis

Symptomatic Radiation Necrosis (mos)
CovariateHR (95% CI)HR p ValueType 3 p Value
Margin size, mm0.028
 >13.07 (1.13–8.34)0.028
 1Ref
GTV, cm30.615
 >150.75 (0.24–2.30)0.615
 ≤15Ref
Technique0.145
 Framed4.26 (0.61–29.83)0.145
 FramelessRef
GPA class0.166
 0–2.50.73 (0.22–2.37)0.599
 31.84 (0.51–6.67)0.356
 3.5–4Ref
No. of fractions0.023
 >10.13 (0.02–0.76)0.023
 1Ref
Conformity index0.05 (0.00–0.56)0.0150.015

Boldface type indicates statistical significance. Number of observations in the original data set = 132. Number of observations used = 104. Backward selection with an α-level of removal of 0.2 was used. The following variables were removed from the model: ECOG, recursive partitioning analysis class, age, number of brain metastases, extent of resection, sex, presence of extracranial disease, and presence of active systemic disease.

Overall Survival

The median OS for all patients was 15.6 months. On MVA, STR and active systemic disease demonstrated a trend toward worse OS. Margin size and SFED were not statistically significant for OS.

Discussion

To our knowledge, this is the first published study to evaluate the impact of a 1.0-mm PTV margin versus a larger margin size on local recurrence and intracranial toxicity in patients with resected brain metastases. Our investigation shows that expanding the PTV margin beyond 1.0 mm does not improve local recurrence, but is associated with a statistically significant higher rate of symptomatic radiation necrosis compared with a 1.0-mm margin.

In 2011, Choi et al.7 prospectively compared outcomes of postoperative SRS patients who were treated with a 2.0-mm target margin compared with a 0-mm margin. They found that adding a margin around the resection cavity leads to improved local control without increasing toxicity. Our 1-year local recurrence rate for all lesions was 14.3%, which is higher than the 1-year local recurrence rate reported by Choi et al.7 of 9.5%. Unlike Choi et al., we used a 1.0-mm margin for the reference patient group. In addition, the median GTV size for our cavities was 11.3 cm3, which is 30% larger than the median GTV size of 8.5 cm3 reported by Choi et al.7 Using spherical target geometry (volume = 4/3πr3), the median cavity size in our study, 11.3 cm3, converts to a median maximum cavity diameter of 2.8 cm. When comparing our overall local recurrence rates to those of other studies, Brennan et al.3 reported a 22% 1-year local recurrence rate with a median maximum cavity diameter of 2.8 cm. Rao et al.25 published in abstract form a 1-year local recurrence rate of 28%, with cavity dimensions reported as percentages: < 2.6 cm, 33%; 2.6–3.5 cm, 42%; and > 3.5 cm, 25%. Next, Soltys et al. reported a 1-year local recurrence rate of 21% after SRS to a median cavity size of 9.8 cm3.28 Thus, despite a larger median cavity size of 11.3 cm3 (or median cavity diameter of 2.8 cm), our 1-year rate of local recurrence appears to be lower when compared with prior reported studies.

Our investigation also shows that cavity size larger than 15 cm3 is a statistically significant predictor for increased rates of local recurrence on MVA, which is a potential reason for why the local recurrence is higher in our study than in that of Choi et al.7 The observation that larger cavity size is associated with local recurrence is congruent with the finding by Minniti et al., where lesion size larger than 3.0 cm was found to be a statistically significant predictor for increased local recurrence in patients with resected melanoma brain metastases.18 To compare our results, a 15-cm3 cavity volume, using spherical geometry, translates to an equivalent lesion diameter of 3.06 cm, which is very similar to the lesion size of 3.0 cm reported by Minniti et al. While Minniti et al.18 only included brain metastases patients with melanoma histology, Rao et al.,25 in a prospective randomized study of resected brain metastases comparing postoperative SRS to observation, found that preoperative tumor size larger than 3.0 cm was associated with higher rates of local recurrence. Collectively, our finding of worse local recurrence with increasing target size is in alignment with that of prior reported studies.

Of the 133 resected brain metastases patients, 16.7% suffered symptomatic radionecrosis at 1 year. Statistically significantly higher 1- and 2-year symptomatic radionecrosis rates were seen in the > 1.0–mm margin cohort than in the 1.0-mm cohort (20.9% and 26.6% vs 6.0% and 9.1%, p = 0.028). Furthermore, all symptomatic patients who required salvage resection were treated with a margin > 1.0 mm (median 1.5 mm). Our overall symptomatic radionecrosis findings are comparable to those of Brennan et al.,3 who reported a radionecrosis rate of 17.5% in 49 patients treated with postoperative radiosurgery. In contrast, Choi et al.7 noted 1-year toxicity rates of only 8%. Our rate of 16.7% is higher than the report by Choi et al. due to a few underlying factors. First, in our study, radionecrosis was diagnosed radiographically and hence could account for higher reported rates of toxicity. Second, target size is a known predictor for toxicity with intracranial SRS. Our median cavity size was 11.3 cm3, whereas the median cavity size as reported by Choi et al.7 was 8.5 cm3. Finally, in our study the median SFED was higher in the > 1.0–mm cohort than in the 1.0-mm cohort (17.5 vs 15 Gy), as was reported by Choi et al. (17.5 vs 16 Gy). Nonetheless, SFED was not significant on MVA for symptomatic radionecrosis.

In support of our findings, Raore et al.26 examined resected margin tissue after GTR of a small group of metastatic lesions from a variety of primary sources. Microscopic assessment of the tissue samples showed no tumor cells infiltrating the surrounding brain tissue. This suggested the need to target only a narrow depth of the resection cavity margin (1.0–2.0 mm) to minimize normal tissue injury and prevent adverse effects from SRS. With regard to other factors that persisted on MVA for symptomatic radionecrosis, number of fractions > 1 was found to be a statistically significant protective variable. This finding is consistent with multiple prior reported studies that support the use of hypofractionated SRS for large brain metastases to mitigate the associated toxicity of radionecrosis.9–11,19

We acknowledge a few pertinent limitations of our study. Radionecrosis was determined radiographically, which could lead to higher apparent rates of toxicity when compared with other studies. All of our patients were treated with LINAC-based SRS, and hence it might make it difficult to extrapolate to Gamma Knife radiosurgery. In addition, dosimetric variables (e.g., V10Gy, V12Gy, V15Gy) were not included in the analysis. Finally, similar to other published studies, we have evaluated multiple histologies, which makes it harder to generalize survival results.

Conclusions

This study demonstrates that increasing the PTV margin beyond 1.0 mm for SRS to resected brain metastases results in no improvement in local control but leads to statistically significant increased rates of symptomatic radionecrosis. Furthermore, in congruence with other published studies, cavity size > 3.0 cm is associated with higher rates of local recurrence.

Acknowledgments

Research reported in this publication was supported in part by the Biostatistics and Bioinformatics Shared Resource of Winship Cancer Institute of Emory University and National Institutes of Health/National Cancer Institute under award no. P30CA138292. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Disclosures

The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

Author Contributions

Conception and design: Patel, Jhaveri. Acquisition of data: Patel, Jhaveri, Chowdhary, Press. Analysis and interpretation of data: Patel, Jhaveri, Press, Dhabaan. Drafting the article: Jhaveri, Chowdhary, Ferris, Morgan, Roper, Elder, Eaton. Critically revising the article: Patel, Jhaveri, Chowdhary, Press, Ferris, Morgan, Roper, Dhabaan, Eaton, Olson, Curran, Shu, Crocker. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Patel. Statistical analysis: Jhaveri, Zhang, Switchenko. Administrative/technical/material support: Roper, Dhabaan, Elder. Study supervision: Patel.

Supplemental Information

Online-Only Content

Supplemental material is available with the online version of the article.

Supplementary Figures and Table. https://thejns.org/doi/suppl/10.3171/2017.9.JNS171735.

References

  • 1

    Asher AL, Burri SH, Wiggins WF, Kelly RP, Boltes MO, Mehrlich M, : A new treatment paradigm: neoadjuvant radiosurgery before surgical resection of brain metastases with analysis of local tumor recurrence. Int J Radiat Oncol Biol Phys 88:899906, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2

    Blonigen BJ, Steinmetz RD, Levin L, Lamba MA, Warnick RE, Breneman JC: Irradiated volume as a predictor of brain radionecrosis after linear accelerator stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 77:9961001, 2010

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 3

    Brennan C, Yang TJ, Hilden P, Zhang Z, Chan K, Yamada Y, : A phase 2 trial of stereotactic radiosurgery boost after surgical resection for brain metastases. Int J Radiat Oncol Biol Phys 88:130136, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4

    Brown PD, Asher AL, Ballman KV, Farace E, Cerhan JH, Anderson SK, : NCCTG N0574 (Alliance): A phase III randomized trial of whole brain radiation therapy (WBRT) in addition to radiosurgery (SRS) in patients with 1 to 3 brain metastases. J Clin Oncol 33 (18 Suppl):LBA4, 2015 (Abstract)

    • Search Google Scholar
    • Export Citation
  • 5

    Brown PD, Ballman KV, Cerhan J, Anderson SK, Carrero XW, Whitton A, : N107C/CEC.3: A Phase III trial of post-operative stereotactic radiosurgery (SRS) compared with whole brain radiotherapy (WBRT) for resected metastatic brain disease. Int J Radiat Oncol Biol Phys 96:937, 2016 (Abstract)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6

    Chang EL, Wefel JS, Hess KR, Allen PK, Lang FF, Kornguth DG, : Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol 10:10371044, 2009

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7

    Choi CY, Chang SD, Gibbs IC, Adler JR, Harsh GR IV, Lieberson RE, : Stereotactic radiosurgery of the postoperative resection cavity for brain metastases: prospective evaluation of target margin on tumor control. Int J Radiat Oncol Biol Phys 84:336342, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8

    Chowdhary M, Patel KR, Danish HH, Lawson DH, Khan MK: BRAF inhibitors and radiotherapy for melanoma brain metastases: potential advantages and disadvantages of combination therapy. Onco Targets Ther 9:71497159, 2016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 9

    Eaton BR, LaRiviere MJ, Kim S, Prabhu RS, Patel K, Kandula S, : Hypofractionated radiosurgery has a better safety profile than single fraction radiosurgery for large resected brain metastases. J Neurooncol 123:103111, 2015 (Erratum in J Neurooncol 123:113, 2015)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10

    Fokas E, Henzel M, Surber G, Kleinert G, Hamm K, Engenhart-Cabillic R: Stereotactic radiosurgery and fractionated stereotactic radiotherapy: comparison of efficacy and toxicity in 260 patients with brain metastases. J Neurooncol 109:9198, 2012

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 11

    Kim YJ, Cho KH, Kim JY, Lim YK, Min HS, Lee SH, : Single-dose versus fractionated stereotactic radiotherapy for brain metastases. Int J Radiat Oncol Biol Phys 81:483489, 2011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12

    Kirkpatrick JP, Wang Z, Sampson JH, McSherry F, Herndon JE II, Allen KJ, : Defining the optimal planning target volume in image-guided stereotactic radiosurgery of brain metastases: results of a randomized trial. Int J Radiat Oncol Biol Phys 91:100108, 2015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 13

    Kocher M, Soffietti R, Abacioglu U, Villà S, Fauchon F, Baumert BG, : Adjuvant whole-brain radiotherapy versus observation after radiosurgery or surgical resection of one to three cerebral metastases: results of the EORTC 22952-26001 study. J Clin Oncol 29:134141, 2011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 14

    Kwon AK, Dibiase SJ, Wang B, Hughes SL, Milcarek B, Zhu Y: Hypofractionated stereotactic radiotherapy for the treatment of brain metastases. Cancer 115:890898, 2009

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 15

    Mahajan A, Ahmed S, Li J, McAleer M, Weinberg J, Brown P, : Postoperative stereotactic radiosurgery versus observation for completely resected brain metastases: results of a prospective randomized study. Int J Radiat Oncol Biol Phys 96 (2 Suppl):S2, 2016 (Abstract)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16

    Mathieu D, Kondziolka D, Flickinger JC, Fortin D, Kenny B, Michaud K, : Tumor bed radiosurgery after resection of cerebral metastases. Neurosurgery 62:817824, 2008

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17

    Minniti G, Clarke E, Lanzetta G, Osti MF, Trasimeni G, Bozzao A, : Stereotactic radiosurgery for brain metastases: analysis of outcome and risk of brain radionecrosis. Radiat Oncol 6:48, 2011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 18

    Minniti G, Paolini S, D’Andrea G, Lanzetta G, Cicone F, Confaloni V, : Outcomes of postoperative stereotactic radiosurgery to the resection cavity versus stereotactic radiosurgery alone for melanoma brain metastases. J Neurooncol 132:455462, 2017

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 19

    Minniti G, Scaringi C, Paolini S, Lanzetta G, Romano A, Cicone F, : Single-fraction versus multifraction (3 × 9 Gy) stereotactic radiosurgery for large (>2 cm) brain metastases: a comparative analysis of local control and risk of radiation-induced brain necrosis. Int J Radiat Oncol Biol Phys 95:11421148, 2016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 20

    Patchell RA, Tibbs PA, Regine WF, Dempsey RJ, Mohiuddin M, Kryscio RJ, : Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA 280:14851489, 1998

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21

    Patel KR, Burri SH, Asher AL, Crocker IR, Fraser RW, Zhang C, : Comparing preoperative with postoperative stereotactic radiosurgery for resectable brain metastases: a multi-institutional analysis. Neurosurgery 79:279285, 2016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 22

    Patel KR, Chowdhary M, Switchenko JM, Kudchadkar R, Lawson DH, Cassidy RJ, : BRAF inhibitor and stereotactic radiosurgery is associated with an increased risk of radiation necrosis. Melanoma Res 26:387394, 2016

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 23

    Patel KR, Lawson DH, Kudchadkar RR, Carthon BC, Oliver DE, Okwan-Duodu D, : Two heads better than one? Ipilimumab immunotherapy and radiation therapy for melanoma brain metastases. Neuro Oncol 17:13121321, 2015

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 24

    Patel KR, Prabhu RS, Kandula S, Oliver DE, Kim S, Hadjipanayis C, : Intracranial control and radiographic changes with adjuvant radiation therapy for resected brain metastases: whole brain radiotherapy versus stereotactic radiosurgery alone. J Neurooncol 120:657663, 2014

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 25

    Rao G, Ahmed S, Hess K, Mahajan A: Postoperative stereotactic radiosurgery vs observation for completely resected brain metastases: results of a prospective randomized study. Neurosurgery 63 (Suppl 1):184, 2016 (Abstract)

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 26

    Raore B, Schniederjan M, Prabhu R, Brat DJ, Shu HK, Olson JJ: Metastasis infiltration: an investigation of the postoperative brain-tumor interface. Int J Radiat Oncol Biol Phys 81:10751080, 2011

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27

    Shaw E, Scott C, Souhami L, Dinapoli R, Bahary JP, Kline R, : Radiosurgery for the treatment of previously irradiated recurrent primary brain tumors and brain metastases: initial report of Radiation Therapy Oncology Group protocol (90-05). Int J Radiat Oncol Biol Phys 34:647654, 1996

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation
  • 28

    Soltys SG, Adler JR, Lipani JD, Jackson PS, Choi CY, Puataweepong P, : Stereotactic radiosurgery of the postoperative resection cavity for brain metastases. Int J Radiat Oncol Biol Phys 70:187193, 2008

    • Crossref
    • PubMed
    • Search Google Scholar
    • Export Citation

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

Contributor Notes

Correspondence Kirtesh R. Patel: Yale School of Medicine, New Haven, CT. kirtesh.patel@yale.edu.

INCLUDE WHEN CITING Published online April 20, 2018; DOI: 10.3171/2017.9.JNS171735.

J.J. and M.C. contributed equally to this work and share first authorship.

Disclosures The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.

  • View in gallery

    Cumulative incidence of local recurrence stratified by margin group. Figure is available in color online only.

  • View in gallery

    Cumulative incidence of symptomatic radiation necrosis stratified by margin group. Figure is available in color online only.

Metrics

All Time Past Year Past 30 Days
Abstract Views 553 153 0
Full Text Views 496 381 34
PDF Downloads 262 163 13
EPUB Downloads 0 0 0