Optimal hypofractionated conformal radiotherapy for large brain metastases in patients with high risk factors: a single-institutional prospective study
© Inoue et al.; licensee BioMed Central Ltd. 2014
Received: 5 August 2014
Accepted: 6 October 2014
Published: 17 October 2014
A single-institutional prospective study of optimal hypofractionated conformal radiotherapy for large brain metastases with high risk factors was performed based on the risk prediction of radiation-related complications.
Eighty-eight patients with large brain metastases ≥10 cm3 in critical areas treated from January 2010 to February 2014 using the CyberKnife were evaluated. The optimal dose and number of fractions were determined based on the surrounding brain volume circumscribed with a single dose equivalent (SDE) of 14 Gy (V14) to be less than 7 cm3 for individual lesions. Univariate and multivariate analyses were conducted.
As a result of optimal treatment, 92 tumors ranging from 10 to 74.6 cm3 (median, 16.2 cm3) in volume were treated with a median prescribed isodose of 57% and a median fraction number of five. In order to compare the results according to the tumor volume, the tumors were divided into the following three groups: 1) 10–19.9 cm3, 2) 20–29.9 cm3 and 3) ≥30 cm3. The lesions were treated with a median prescribed isodose of 57%, 56% and 55%, respectively, and the median fraction number was five in all three groups. However, all tumors ≥20 cm3 were treated with ≥ five fractions. The median SDE of the maximum dose in the three groups was 47.2 Gy, 48.5 Gy and 46.5 Gy, respectively. Local tumor control was obtained in 90.2% of the patients, and the median survival was nine months, with a median follow-up period of seven months (range, 3-41 months). There were no significant differences in the survival rates among the three groups. Six tumors exhibited marginal recurrence 7-36 months after treatment. Ten patients developed symptomatic brain edema or recurrence of pre-existing edema, seven of whom required osmo-steroid therapy. No patients developed radiation necrosis requiring surgical resection.
Our findings demonstrate that the administration of optimal hypofractionated conformal radiotherapy based on the dose-volume prediction of complications (risk line for hypofractionation), as well as Kjellberg’s necrosis risk line used in single-session radiosurgery, is effective and safe for large brain metastases or other lesions in critical areas.
Surgical removal is the gold standard therapy and is essential for treating large brain metastases causing progressive symptoms due to increased intracranial pressure, as the symptoms improve immediately after surgery. However, surgical removal carries a risk of causing neurological deficits after dissecting critical areas, especially in cases of tumors situated deep within the white matter. Surgery also requires hospitalization for at least one week, with higher medical expenses than that observed for radiosurgery in this country. In addition, there are many patients with general risks for surgery, as well as those who refuse surgical procedures, due to having primary malignancies and/or a poor performance status.
Whole brain radiotherapy (WBRT) and chemotherapy are not adequate to control large brain metastases, and radiosurgery is an important therapeutic tool for treating brain metastases in multiple clinical settings. Moreover, radiosurgery is increasingly being used as a primary treatment modality in an attempt to prevent disturbances in the neurocognitive function after WBRT . However, single-session radiosurgery is also inadequate for managing large brain metastases due to dose limitations resulting from the need to prevent adverse effects on the surrounding structures, such as the optic pathway, internal capsule and brainstem .
Hypofractionated radiotherapy appears to be beneficial in case of metastases not causing clinical signs of impending cerebral herniation, and its use is supported by the findings of previously published series employing varying radiation dose and fractionation schedules –. However, the optimal dose and number of fractions have yet to be established , and the exact incidence of adverse effects on the surrounding brain is unclear in patients with tumors with high risk factors, such as a large size or location in a critical area. Therefore, in order to determine the optimal dose and fractionation schedule, dose escalation following low-dose treatment is required.
We previously reported that the brain volume circumscribed with a single dose equivalent (SDE) of 14 Gy (V14) is an indicator of radiation necrosis  and that the incidence of radiation-related complications after hypofractionated conformal radiotherapy is best predicted according to the dose-volume relationship using the SDE of the maximum dose and V14 . In January 2010, we initiated a prospective study of optimal hypofractionated conformal radiotherapy based on risk prediction in order to avoid radiation necrosis after hypofractionation treatment for large brain metastases in addition to the Kjellberg’s necrosis risk line used in single-session radiosurgery .
This report presents the results of our single-institution prospective study of optimal hypofractionated conformal radiotherapy performed in this institute as useful treatment for patients with large brain metastases in critical areas and/or those with general risks for surgery.
All patients provided their written informed consent prior to the procedure with institutional ethics committee approval. Ninety-seven patients with large brain metastases measuring 10 cm3 in volume or more were treated with optimal hypofractionated radiotherapy based on the dose-volume prediction of complications in order to avoid radiation necrosis and subsequently followed more than three months at Kanto Neurosurgical Hospital between January 2010 and February 2014. An evaluation of the rates of local tumor control, overall survival and complications was performed as a prospective single-institutional analysis.
Definition of variables and end points
The treatment dose is expressed as the marginal dose used in hypofractionated conformal radiotherapy. The maximum dose was automatically obtained from the marginal dose and the prescribed isodose delivered to the lesion margin. A complication was defined as neurological impairment (either the development of a new deficit or the significant deterioration of a preexisting or recurrent deficit) with a change on either computerized tomography scans or magnetic resonance imaging studies. Clinical follow-up was considered to have stopped at the time of the most recent report from the patient and/or a representative or the time of death.
Inclusion and exclusion criteria
Ninety-seven patients followed for more than three months after treatment were found to be eligible for inclusion in this study. Nine patients (9.3%) were excluded from the analysis due to a lack of available imaging findings after treatment. All other patients (88) were included in the analysis.
Pretreatment characteristics of the 88 patients with large brain metastases in critical areas
Number of patients
Location of tumor
Median age (range)
Thalamus, basal ganglia
Median KPS score
Tumor volume, median (cm3)
Image follow-up period (months)
Multiple vs single
Survival period (months)
Metastases to other organs
Prescribed marginal and SDE of the maximum dose
The maximum dose was calculated based on the marginal dose and prescribed isodose. For the purpose of the dose-volume analysis, the maximum dose in three- to ten-fraction treatment was converted to the SDE using the equation reported by Park et al. and Eaton et al. ,, as previously reported .
Optimal hypofractionated conformal radiotherapy
Hypofractionated conformal radiotherapy was administered under CT and MRI guidance as previously reported . When setting a dose and fraction schedule as the first plan, a marginal dose of 27–30 Gy in three fractions was intended to use to treat tumors measuring 10–19.9 cm3. A marginal dose of 31–35 Gy in five fractions and a marginal dose of 35–42 Gy in eight to 10 fractions were intended to use to treat tumors measuring 20–29.9 cm3 and tumors measuring 30 cm3 or more, respectively. The isodose volume of the surrounding brain (excluding the GTV) circumscribed with an SDE of 14 Gy (V14), as well as the tumor volume, was measured using the MultiPlan software program for the G4 system (Accuray, Sunnyvale, CA) and recorded in each patient in order to obtain the optimal dose and fractionation schedule. The SDE of 14 Gy used in three- to ten-fraction treatment was 23.1 to 38.4 Gy according to Timmerman’s values , as previously reported . The V14 of each tumor was maintained at less than 7 cm3 in order to prevent radiation necrosis for optimal treatment based on the dose-volume prediction of complications . If the V14 value was more than 7 cm3 in the first dose-plan, it was lowered to less than 7 cm3 using a decreased marginal isodose or dose or an increased number of fractions in the revised or re-revised dose-plan, maintaining an effective marginal dose (SDE: 18–20 Gy) on the target.
Follow-up evaluations and complications
Changes in the patients’ neurological symptoms, such as paresis, sensory disturbances, aphasia or visual disturbances, were examined after treatment. Serial imaging studies (MRI or CT) were requested six weeks after treatment and every two to three months thereafter. Symptomatic brain edema was identified in association with neurological deterioration and imaging changes in the concomitant perifocal regions after treatment. The tumor response was divided into three groups, reduced (a decrease in the tumor volume of >15%), stable (a change in the tumor volume within ±15%) or enlarged (an increase in the tumor volume of >15%), in order to compare the response with that observed in former reports of large brain metastases ,. The incidence of complications was examined in relation to the V14 of the surrounding brain.
Univariate and multivariate analyses were conducted using logistic regression and Cox hazard models. Differences between the groups were evaluated using Student’s t-test. Overall survival was estimated according to the Kaplan-Meier method and examined for significance using the log-rank and generalized Wilcoxon tests. All analyses employed the conventional p < 0.05 level of significance.
Nineteen patients received osmo-steroid therapy during hypofractionation treatment for symptoms of perifocal edema and/or the further oral administration of steroids depending on the presence of other symptoms at the outpatient clinic.
No new neurological deficits from direct damage to the optic pathway, brainstem or functional areas were noted, although symptoms recurred or appeared in 10 patients due to adverse effects (brain edema) on the surrounding brain.
Treatment-related variables of optimal hypofractionation
Treatment-related variables of the 92 large brain metastases in the 88 patients
Prescribed isodose (%)
3 - 10
Lesion treated with 3 fraction
Maximum dose (single fraction equivalent dose, Gy)
Results of optimal hypofractionation in the three groups divided according to the tumor volumes
Tumor volume (cm3) (median)
Median isodose (%) (range)
Median fraction Number (range)
Median SDE of the max. dose (Gy) (range)
Median V14 (cm3)
Marginal recurrences (6)
Complications (10) (brain edema)
Tumor response, local control and overall survival after treatment
Characteristics of the six patients with recurrences after optimal hypofractionated conformal radiotherapy
M: 3, F: 3
Cerebrum (P: 2, F: 2, T: 1, O: 1)
Tumor volume (cm3)
Prescribed isodose (%)
SDE of the max. dose (Gy)
Adverse effects (brain edema)
Characteristics of the 10 patients with adverse effects after optimal hypofractionated conformal radiotherapy
p value (CH model)
M: 7, F: 3
Cerebrum: 9, cerebellum: 1
Tumor volume (cm3)
Prescribed isodose (%)
SDE of the max. dose (Gy)
The prognosis of patients with brain metastases is related to the stage of the primary cancer, age and the KPS score ,. The worst survival is seen in patients with a KPS of less than 70. In the present series, 38.6% of the patients had a KPS of less than 70 and 65.9% of the patients were 60 years of age or older. Although 78 patients (88.6%) were in RTOG-RPA class 2 or 3, the median survival of our patients was nine months. Furthermore, no statistically significant differences were found between the three groups of patients divided according to the tumor volume, although the survival rate and median KPS score were lowest in the largest group. Optimal hypofractionated conformal radiotherapy helps to increase the KPS, at least in patients with symptomatic lesions not directly affecting functional areas, and contributes to improving the prognosis of patients with large brain metastases, as previously reported .
Single-session radiosurgery is increasingly being used to treat brain metastases and has the benefits of a short treatment time, high tumor control rate and low risk of complications. However, large metastases are not suitable for treatment with single-session radiosurgery, as lower tumor control rates (85%) and higher complication rates (15%) than those for smaller metastases have been reported ,.
The optimal hypofractionation treatment in this series yielded a tumor control rate of 90.2% in the patients with large tumors. The median maximum dose (SDE) of 47.4 Gy at a median prescribed isodose of 57% in a median of five fractions appeared to be effective for most large brain metastases, in addition to a marginal dose of 20 Gy at the prescribed isodose of 50-60% for small tumors in single fraction radiosurgery. However, more than five fractions were used in cases involving large tumors measuring more than 20 cm3 or tumors associated with extensive brain edema in order to decrease the V14 values.
Consequently, tumor recurrence appeared in six patients more than six months after treatment, all of which originated from marginal areas treated with the prescribed isodose. Additional treatment was easily performed in these patients, because the volume of the recurrent tumors was not large, and the risk of radiation necrosis after the second treatment was assessed to be very low. As to risk factors for recurrence, a larger tumor volume, lower prescribed isodose and lower SDE of the maximum dose are potential candidates; however, no factors were found to be significant in the univariate or multivariate analyses in this study. The difference in tumor volume (10–19.9 cm3 vs 30–74.6 cm3) between the groups was significant; however, recurrence occurred in only four of 61 patients with smaller tumors. Although greater sample size is required for further statistical analyses, our findings indicate that optimal hypofractionated conformal radiotherapy is effective for treating large lesions with a low rate of recurrence.
Brain edema developed in 10 patients, mostly within six months. With respect to risk factors, age and the number of fractions were found to be significant in this series. Older patients’ brains with large metastases may be sensitive to irradiation stress or possibly exhibit greater vulnerability than the normal adult brain. More than five fractions were used to treat large tumor measuring more than 20 cm3, tumors associated with extensive edema or tumors located in critical areas. Tumors situated deep within the white matter have a tendency to cause brain edema, and factors related to the onset of edema may promote the development or re-appearance of brain edema. In contrast to that observed for radiation necrosis, most cases of brain edema developed several months after treatment, and all were reversible and recovered after either osmo-steroid therapy or oral steroid treatment. Therefore, optimal hypofractionated conformal radiotherapy is a safe treatment for patients with large metastases with high risk factors.
Conducting dose-volume prediction of complications is essential for providing optimal hypofractionation treatment. We previously reported a method for predicting radiation necrosis using a model that accounted for the SDE of the maximum dose and V14 . Long-term experience with single-session radiosurgery has also confirmed the optimal treatment doses for individual pathologies in the brain. For example, a marginal dose of 12 Gy or 20 Gy is used to treat vestibular schwannoma or AVM, respectively. The long-term results have been shown to be satisfactory, with low rates of complications ,. The optimal dose and fraction number for hypofractionation treatment for such pathologies may be determined by predicting the incidence of complications with respect to avoiding adverse effects on the surrounding brain.
In this prospective study, our findings demonstrated a rate of high tumor control and a low rate of complications in the treatment of large brain metastases with high risk factors. No patients with radiation necrosis required surgical resection during the median follow-up of seven to 41 months. Therefore, the administration of defined optimal hypofractionation treatment based on the dose-volume prediction of complications is effective and safe for the treatment of large lesions. However, the V14 may need to be further reduced to less than 3 cm3 when treating tumors situated deep within the white matter and/or exhibiting extensive perifocal edema, as the rate of recurrence of pre-existing edema was not low in the patients with a V14 of 3 cm3 or more after treatment in this series. Conformal radiotherapy with a prescribed isodose of 50% to 60% has the benefits of decreasing the V14 value due to a sharp fall-off in the dose distribution, as noted with single-session radiosurgery. Optimal treatment using a large fraction number also has the benefits of decreasing the V14 value and helps to avoid radiation necrosis, as demonstrated in this prospective study.
This prospective study of optimal dose and fractionation treatment for large brain metastases with high risk factors showed satisfactory results for local tumor control and survival, with limited complications. Conducting dose-volume prediction of complications using the V14 is beneficial for preventing irreversible complications in the treatment of large brain metastases or other lesions in the brain.
HKI developed the concept of study, contributed to the collection and analysis of the data and wrote the manuscript, SH, SY, SJ and NS conducted the treatment planning, SK and TK contributed to administering the treatment and supported the data analysis, SH and NT provide administrative support. All authors approved the final manuscript.
- Chang EL, Wefel JS, Hess KR, Allen PK, Lang FF, Kornguth DG, Arbuckle RB, Swint JM, Shiu AS, Maor MH, Meyers CA: Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol. 2009, 10: 1037-1044. 10.1016/S1470-2045(09)70263-3.View ArticlePubMedGoogle Scholar
- Lee CC, Yen CP, Xu Z, Schlesinger D, Sheehan J: Large intracranial metastatic tumors treated by Gamma Knife surgery: outcomes and prognostic factors. J Neurosurg. 2014, 120: 52-59. 10.3171/2013.9.JNS131163.View ArticlePubMedGoogle Scholar
- Aoyama H, Shirato H, Onimaru R, Kagei K, Ikeda J, Ishii N, Sawamura Y, Miyasaka K: Hypofractionated stereotactic radiotherapy alone without whole-brain irradiation for patients with solitary and oligo brain metastasis using noninvasive fixation of the skull. Int J Radiat Oncol Biol Phys. 2003, 56: 793-800. 10.1016/S0360-3016(03)00014-2.View ArticlePubMedGoogle Scholar
- Lindvall P, Bergström P, Löfroth PO, Henriksson R, Bergenheim AT: Hypofractionated conformal stereotactic radiotherapy alone or in combination with whole-brain radiotherapy in patients with cerebral metastases. Int J Radiat Oncol Biol Phys. 2005, 61: 1460-1466. 10.1016/j.ijrobp.2004.08.027.View ArticlePubMedGoogle Scholar
- Marchetti M, Milanesi I, Falcone C, De Santis M, Fumagalli L, Brait L, Bianchi L, Fariselli L: Hypofractionated stereotactic radiotherapy for oligometastases in the brain: a single-institution experience. Neurol Sci. 2011, 32: 393-399. 10.1007/s10072-010-0473-4.View ArticlePubMedGoogle Scholar
- Ogura K, Mizowaki T, Ogura M, Sakanaka K, Arakawa Y, Miyamoto S, Hiraoka M: Outcomes of hypofractionated stereotactic radiotherapy for metastatic brain tumors with high risk factors. J Neurooncol. 2012, 109: 425-432. 10.1007/s11060-012-0912-6.View ArticlePubMedGoogle Scholar
- Märtens B, Janssen S, Werner M, Frühauf J, Christiansen H, Bremer M, Steinmann D: Hypofractionated stereotactic radiotherapy of limited brain metastases: a single-centre individualized treatment approach. BMC Cancer. 2012, 12: 497-10.1186/1471-2407-12-497.PubMed CentralView ArticlePubMedGoogle Scholar
- Eaton BR, Gebhardt B, Prabhu R, Shu HK, Curran WJ, Crocker I: Hypofractionated radiosurgery for intact or resected brain metastases: defining the optimal dose and fractionation. Radiat Oncol. 2013, 8: 135-10.1186/1748-717X-8-135.PubMed CentralView ArticlePubMedGoogle Scholar
- Inoue HK, Sato H, Seto K, Torikai K, Suzuki Y, Saitoh J, Noda SE, Nakano T: Five-fraction CyberKnife radiotherapy for large brain metastases in critical areas: impact on the surrounding brain volumes circumscribed with a single dose equivalence of 14 Gy (V14) to avoid radiation necrosis. J Radiat Res. 2014, 55: 334-342. 10.1093/jrr/rrt127.PubMed CentralView ArticlePubMedGoogle Scholar
- Inoue HK, Sato H, Suzuki Y, Saitoh J, Noda S, Seto K, Torikai K, Nakano T: Dose-volume prediction of radiation-related complications after hypofractionated conformal radiotherapy for brain metastases in critical areas. Cureus. 2014, 6 (7): e189-doi:10.7759/cureus.189Google Scholar
- Barker FG, Butler WE, Lyons S, Cascio E, Ogilvy CS, Loeffler JS, Chapman PH: Dose-volume prediction of radiation-related complications after proton beam radiosurgery for cerebral arteriovenous malformations. J Neurosurg. 2003, 99: 254-263. 10.3171/jns.2003.99.2.0254.View ArticlePubMedGoogle Scholar
- Park C, Papiez L, Zhang S, Story M, Timmerman RD: Universal survival curve and single fraction equivalent dose: useful tools in understanding potency of ablative radiotherapy. Int J Radiat Oncol Biol Phys. 2008, 70: 847-852. 10.1016/j.ijrobp.2007.10.059.View ArticlePubMedGoogle Scholar
- Timmerman RD: An overview of hypofractionation and introduction to this issue of seminars in radiation oncology. Semin Radiat Oncol. 2008, 18: 215-222. 10.1016/j.semradonc.2008.04.001.View ArticlePubMedGoogle Scholar
- Gaspar L, Scott C, Rotman M, Asbell S, Phillips T, Wasserman T, McKenna WG, Byhardt R: Recursive partitioning analysis (RPA) of prognostic factors in three Radiation Therapy Oncology Group (RTOG) brain metastases trials. Int J Radiat Oncol Biol Phys. 1997, 37: 745-751. 10.1016/S0360-3016(96)00619-0.View ArticlePubMedGoogle Scholar
- Andrews DW, Scott CB, Sperduto PW, Flanders AE, Gaspar LE, Schell MC, Werner-Wasik M, Demas W, Ryu J, Bahary JP, Souhami L, Rotman M, Mehta MP, Curran WJ: 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. 2004, 363: 1665-1672. 10.1016/S0140-6736(04)16250-8.View ArticlePubMedGoogle Scholar
- Williams BJ, Suki D, Fox BD, Pelloski CE, Maldaun MV, Sawaya RE, Lang FF, Rao G: Stereotactic radiosurgery for metastatic brain tumors: a comprehensive review of complications. J Neurosurg. 2009, 111: 439-448. 10.3171/2008.11.JNS08984.View ArticlePubMedGoogle Scholar
- Inoue HK: Low-dose radiosurgery for large vestibular schwannomas: long-term results of functional preservation. J Neurosurg. 2005, 102 (Suppl): 111-113. 10.3171/jns.2005.102.s_supplement.0111.View ArticlePubMedGoogle Scholar
- Inoue HK: Long-term results of Gamma Knife surgery for arteriovenous malformations: 10- to 15-year follow up in patients treated with lower doses. J Neurosurg. 2006, 105 (Suppl): 64-68.PubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.