Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions

Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions

G Model CANRAD-3959; No. of Pages 7 ARTICLE IN PRESS Cancer/Radiothérapie xxx (2019) xxx–xxx Disponible en ligne sur ScienceDirect www.sciencedirec...

726KB Sizes 0 Downloads 32 Views

G Model CANRAD-3959; No. of Pages 7

ARTICLE IN PRESS Cancer/Radiothérapie xxx (2019) xxx–xxx

Disponible en ligne sur

ScienceDirect www.sciencedirect.com

Original article

Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions Radiothérapie en conditions stéréotaxiques hypofractionnée adaptée : 36 Gy en six fractions pour les métastases cérébrales critiques D. Dumont Lecomte a,∗ , J. Lequesne b , J. Geffrelot c , P. Lesueur c , V. Barraux d , C. Loiseau d , J. Lacroix e , A. Leconte b , É. Émery f,g , J. Thariat c,g,h,1 , D. Stefan c,1 a

Radiothérapie, CHU de Bordeaux hôpital Haut-Lévêque, avenue de Magellan, 33604 Pessac, France Recherche clinique, centre Franc¸ois-Baclesse, 3, avenue du Général-Harris, 14000 Caen, France c Radiothérapie, centre Franc¸ois-Baclesse, 3, avenue du Général-Harris, 14000 Caen, France d Physique médicale, centre Franc¸ois-Baclesse, 3, avenue du Général-Harris, 14000 Caen, France e Radiologie, centre Franc¸ois-Baclesse, Caen, 3, avenue du Général-Harris, 14000 Caen, France f Neurochirurgie, CHU de Caen, avenue de la Côte-de-Nacre, 14000 Caen, France g Université de Caen Normandie, esplanade de la Paix, 14003 Caen, France h Laboratoire de physique corpusculaire IN2P3/ENSICAEN - UMR6534, Unicaen, 6, boulevard du Maréchal-Juin, 14000 Caen, France b

a r t i c l e

i n f o

Article history: Received 7 May 2018 Received in revised form 10 June 2019 Accepted 13 June 2019 Keywords: Haemorrhage Brain metastasis Hypo fractionated Radiation therapy Radiotherapy Radiosurgery

a b s t r a c t Purpose. – Stereotactic radiosurgery and hypofractionated stereotactic radiotherapy are standard treatments for brain metastases when they are small in size (at the most 3 cm in diameter) and limited in number, in patients with controlled extracerebral disease and a good performance status. Large inoperable brain metastases usually undergo hypofractionated stereotactic radiotherapy while haemorrhagic brain metastases have often been contraindicated for both stereotactic radiosurgery or hypofractionated stereotactic radiotherapy. The objective of this retrospective study was to assess a six 6 Gy-fractions hypofractionated stereotactic radiotherapy scheme in use at our institution for haemorrhagic brain metastases, large brain metastases (size greater than 15 cm3 ) or brain metastases located next to critical structures. Material and methods. – Patients with brain metastases treated with the 6 × 6 Gy scheme since 2012 to 2016 were included. Haemorrhagic brain metastases were defined by usual criteria on CT scan and MRI. Efficacy, acute and late toxicity were evaluated. Results. – Sixty-two patients presenting 92 brain metastases were included (32 haemorrhagic brain metastases). Median follow up was 10.1 months. One-year local control rate for haemorrhagic brain metastases, large brain metastases, or brain metastases next to critical structures were 90.7%, 73% and 86.7% respectively. Corresponding overall survival rates were 61.2%, 32% and 37.8%, respectively. Haemorrhagic complications occurred in 5.3% of patients (N = 5), including two cases of brain metastases with pretreatment haemorrhagic signal. Tolerance was good with only one grade 3 acute toxicity. Conclusion. – The 6 × 6 Gy hypofractionated stereotactic radiotherapy scheme seems to yield quite good results in patients with haemorrhagic brain metastases, which must be confirmed in a prospective way. ´ e´ franc¸aise de radiotherapie ´ oncologique (SFRO). Published by Elsevier Masson SAS. All © 2019 Societ rights reserved.

r é s u m é Mots clés : Hémorragie Métastase cérébrale

Objectif de l’étude. – La radiochirurgie et la radiothérapie stéréotaxique sont le standard de traitement des métastases cérébrales lorsqu’elles sont peu nombreuses, de petite taille, chez les patients en bon état général et à la maladie extracérébrale contrôlée. Les métastases plus volumineuses sont souvent

∗ Auteur correspondant. E-mail address: [email protected] (D. Dumont Lecomte). 1 Equally contributed to the manuscript. https://doi.org/10.1016/j.canrad.2019.06.012 ´ e´ franc¸aise de radiotherapie ´ 1278-3218/© 2019 Societ oncologique (SFRO). Published by Elsevier Masson SAS. All rights reserved.

Please cite this article in press as: Dumont Lecomte D, et al. Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions. Cancer Radiother (2019), https://doi.org/10.1016/j.canrad.2019.06.012

G Model CANRAD-3959; No. of Pages 7

ARTICLE IN PRESS D. Dumont Lecomte et al. / Cancer/Radiothérapie xxx (2019) xxx–xxx

2 Hypofractionné Radiothérapie Stéréotaxie Radiochirurgie

acceptées en stéréotaxie, tandis que les métastases hémorragiques restent plutôt contre-indiquées pour ce type de traitement. L’objectif de cette étude rétrospective était de rapporter les résultats obtenus par un schéma de fractionnement 6 × 6 Gy utilisé dans notre centre pour les métastases hémorragiques, de plus de 15 cm3 ou situées proches de structures à risque. Matériel et méthodes. – Les patients traités par ce schéma dans notre centre de 2012 à 2016 ont été inclus. Le caractère hémorragique était défini sur l’imagerie scanographique et IRM. L’efficacité et la tolérance ont été évaluées. Résultats. – Soixante-deux patients atteints de 92 métastases (32 hémorragiques) ont été inclus. Le suivi médian était de 10,1 mois. Le taux de contrôle local à 1 an pour les métastases hémorragiques, les grosses métastases, et les métastases proches de structure à risque étaient respectivement de 90,7 %, 73 % et 86,7 %. Les taux correspondants de survie globale étaient de 61,2 %, 32 % et 37,8 %. Des complications hémorragiques ont été observées chez 5,3 % des patients (deux lésions hémorragiques avant le traitement). Un seul cas de toxicité aiguë de grade 3 a été observé, résolutif après traitement symptomatique. Conclusion. – Le Schéma 6 × 6 Gy semble réalisable et plutôt efficace en traitement des métastases hémorragiques, ce qui reste à confirmer en prospectif. ´ e´ franc¸aise de radiotherapie ´ oncologique (SFRO). Publie´ par Elsevier Masson SAS. Tous © 2019 Societ ´ ´ droits reserv es.

1. Introduction Brain metastases occur in about one third of cancer patients, especially in lung cancer and melanoma [1]. The decision making process toward palliation, whole brain radiation therapy, surgery, radiosurgery (using high dose single fraction stereotactic irradiation) or hypofractionated stereotactic irradiation is based on life expectancy, symptoms, controlled primary, brain metastases characteristics (number, size and location) and prognostic scoring classifications [2,3]. Treatments of brain metastases have evolved over the past 20 years and stereotactic irradiation has become a standard of care in patients with oligometastatic brain disease [4]. Series of brain metastases treated by stereotactic irradiation report local control rates of 69 to 81% at one year for both stereotactic radiosurgery or hypofractionated stereotactic radiotherapy [5–8]. Compared to whole brain radiation therapy, both techniques of radiosurgery and hypofractionated stereotactic radiotherapy have the advantage of less neurocognitive sequelae with similar overall survival [9]. However, under such circumstances of large tumour size and proximity of functional areas, brain metastases at risk for functional deterioration after radiosurgery may benefit from hypofractionated stereotactic irradiation [8,10–12]. Haemorrhagic brain metastases are often contraindicated for radiosurgery due to the belief that haemorrhage at diagnosis is associated with poorer prognosis, because harder to delineate and often bigger than others, and with a risk of severe, sometimes lethal, haemorrhagic complications. No specific recommendations are made in this situation [13]. Fractionation may improve the tolerance of stereotactic irradiation in such context. Various fractionation schedules have been designed based on brain metastases size and proximity to critical structures, with the objective of a sufficient biological equivalent dose [14,15]. At our institution, we use three fractionation regimens: a single fraction of 20 to 24 Gy for brain metastases smaller than 1 cm, 3 × 10 Gy for brain metastases from 1 to 3 cm, and 6 × 6 Gy for other critical situations. The objective of this study was to evaluate 6 × 6 Gy hypofractionated stereotactic irradiation in terms of efficacy and tolerance in patients with haemorrhagic, large or brain metastases situated next to critical structures. 2. Material and methods 2.1. Patient and treatment characteristics In agreement with the French law and Commission nationale informatique et libertés (Cnil), patients were informed of the study

and retrospective data were anonymized and stored on a secured server. The study was approved by an ethics committee. All consecutive adult patients treated at our institution for one or more brain metastases using CyberKnifeTM stereotactic irradiation using 6 × 6 Gy between 2012 and end of 2016 were included. Patients who had received previous whole brain radiation therapy or surgical resection of brain metastases were excluded. Each case was discussed in multidisciplinary meeting. The sample included three main groups of indications. Haemorrhagic brain metastases were defined by the presence of a haemorrhagic signal according to Zhang et al. as either a spontaneous (without contrast enhancement) hyperdensity on CT scan, or a spontaneous hypersignal in T1 MRI sequence, or hyposignal in T2* MRI sequence, accepting that in some cases it may correspond to melanin signal [16]. Large brain metastases were defined by a volume greater than 15 cm3 . Brain metastases next to a critical structure were defined as within 1 cm of the optical pathway, the brainstem, the choroid plexus, or in the cerebellum with a risk of cerebral herniation. In other cases, fractionation was chosen because of a large peritumoral oedema, risking cerebral herniation. During the radiotherapy course, a 1 mg/kg oral steroid treatment was given to all patients, in the absence of contraindication. If multiple brain metastases were present, treatments of any other brain metastases were recorded. Systemic treatment (chemotherapy, targeted therapy or immunotherapy) during radiotherapy was recorded. Controlled extra-cerebral disease was defined as stable or regressive on the last CT scan evaluation prior to irradiation of brain metastases. After treatment, patients were followed until the end of the data collection (November 2017) or their death by a clinical evaluation and an MRI every 3 months. Patients without at least one MRI post treatment evaluation were excluded. 2.2. Treatment planning All patients were immobilized using a tight thermoplastic stereotactic head mask. One-millimetre thick CT scan images were obtained for each patient and fused with MRI images. Gross tumour volume was defined as the enhanced volume on T1 weighted contrast enhanced sequences. A 2 mm three-dimensional expansion was applied to the gross tumour volume to create the planning target volume [11,17,18]. Hypofractionated stereotactic radiotherapy was delivered with the CyberKnife® system (Accuray, Sunnyvale) using the skull-tracking module at a dose of 36 Gy in six equallyweighted fractions, every other day. The dose was prescribed on the 80% isodose line.

Please cite this article in press as: Dumont Lecomte D, et al. Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions. Cancer Radiother (2019), https://doi.org/10.1016/j.canrad.2019.06.012

G Model CANRAD-3959; No. of Pages 7

ARTICLE IN PRESS D. Dumont Lecomte et al. / Cancer/Radiothérapie xxx (2019) xxx–xxx

3

2.3. Follow up and statistical analysis

3.2. Efficacy

The primary endpoint was efficacy evaluated by local control for each lesion at 6 and 12 months. Local progression was defined according to the modified Recist 1.1 criteria on MRI. Local control was then defined as any lesion that did not meet criteria for local progression. Local control was estimated through a competing risks methodology, the death being the event competing with local progression of brain metastases. The relationship between clinical and demographic covariates and local control was then estimated using a Fine and Gray competing risks regression model. Descriptive statistics was based on mean, median and range for continuous variables, and on frequencies for categorical variables. Survival probability was estimated using the Kaplan–Meier method. Log-rank tests were performed to evaluate the effects of various variables on survival data. Patients carrying at least one brain metastasis with the following characteristics: large, haemorrhagic or next to a critical structure, were categorized in the corresponding subgroup. Patients were considered in the haemorrhagic subgroup if they had at least that characteristic, regardless of size or functional area. Intracranial progression-free survival was defined as the time from treatment to any intracerebral (local or regional) relapse. Haemorrhagic complication was defined by the appearance of a haemorrhagic signal as previously described, or the significant increase in size of a previous haemorrhagic signal. Salvage treatments, such as surgery, new course of hypofractionated stereotactic radiotherapy or whole brain radiation therapy were also evaluated. Multimodal MRI were performed in case of suspicion of radionecrosis on surveillance MRI. Acute and late toxicity occurring respectively within or after 90 days from the initiation of hypofractionated stereotactic radiotherapy were recorded retrospectively according to the Radiotherapy Oncology Groupt (RTOG) central nervous system toxicity criteria.

Median follow-up was 10.1 months and median survival was 12.2 months. The local control rates at treated sites at 6 months and 1 year were 95.6% and 88.3%, respectively. Overall survival rates at 6 months and 1 year was 77.0% and 51.7%, respectively (Fig. 1). In the haemorrhagic brain metastases subgroup, the local control rates at 6 months and one year were 93.8% and 90.6% respectively, and 96.6% and 86.9% in non-haemorrhagic lesions (either large or near critical structures lesions). Overall survival rates at 6 months and one year were 81.0 and 61.2% in patients presenting at least one haemorrhagic lesion, and 75.0% and 46.9% in other cases (Fig. 2). In the large brain metastases subgroup, local control rates at 6 months and one year were 90% and 80% respectively. Overall survival rates in patients presenting at least one large brain metastasis were 60% and 40% respectively. Volume of brain metastases was significantly associated with haemorrhage. In the subgroup of brain metastases located next to critical structures, the local control rates at 6 months and one year were 100% and 86.7% respectively. Overall survival rates at 6 months and one year in patients with at least one brain metastasis located next to a critical structure were 87.5% and 37.5%, respectively. Intracranial progression-free survival rates were 53% at 6 months and 36.5% at one year. Salvage treatment was necessary in 28 patients, 12 with whole brain radiation therapy, 13 with another course of hypofractionated stereotactic radiotherapy, and three with an association of neurosurgery and another course of postoperative hypofractionated stereotactic radiotherapy. The introduction of a salvage treatment for 28 patients was not significantly associated with survival, with 1-year overall survival rate of 53.6% for these patients versus 50.2% for others (p = 0.76). Performance status 2 or above and age at least 66 years were the only significant prognostic factors decreasing survival (with respectively P = 0.02 and P = 0.03). 3.3. Toxicity

3. Results 3.1. Population and brain metastases characteristics Sixty-three patients (out of 68 identified in the database, i.e. five patients with no follow up MRI available) were treated for 92 brain metastases using the 6 × 6 Gy regimen between 2012 and late 2016. Performance status of patients were mainly 0 or 1, respectively 21 and 60%. Most represented primary lesions were lung cancer, kidney cancer, melanoma and breast cancer. Thirty-one patients had symptomatic brain metastases, which was neither associated with haemorrhage (P = 0.48), nor with a next to critical structure location (P = 1), nor with a large volume (P = 0.21). Patient characteristics are described in Table 1. Eighteen patients had a concomitant treatment (conventional chemotherapy, immunotherapy) and ten patients interrupted their concomitant treatment because of potential cumulative toxicity, mostly with targeted therapies. The 6 × 6 fractionation scheme was chosen because of haemorrhagic brain metastases for 32 lesions in 22 patients, large size for 14 lesions in 14 patients, and location near critical anatomic areas for 16 lesions in 13 patients. Critical locations were the cerebellum with a risk of herniation for ten lesions, adjacent to brainstem in one, to plexus choroid in three, to optical pathway in two, to pineal gland in 1. In the remaining cases, the 6 × 6 fractionation was chosen because of large peritumoral oedema. Brain metastases characteristics are described in Table 2. For the subgroup of “large brain metastases”, mean gross tumour volume was 23.5 cm3 (range: 15.1;42.1 cm3 ). Mean gross tumour volume of brain metastases situated in critical location was 4.8 cm3 (range: 0.3; 25.3 cm3 ).

Acute grade 1 or 2 toxicity occurred in 44.4% of the patients, and mostly consisted of headaches, asthenia and seizure recurrence (N = 2). Only one grade 3 acute toxicity occurred during the course of hypofractionated stereotactic radiotherapy with de novo hemiparesis. This patient had significant peri tumoral oedema before hypofractionated stereotactic radiotherapy and recovered quickly under intra-venous anti-oedematous treatment. For late toxicity, more follow up is needed, but no grade 3 or 4 have yet occurred. Haemorrhagic complications occurred in five lesions, two of them being already haemorrhagic before the radiotherapy course. There was no dominant histology. In only one case this haemorrhagic complication turned into a neurologic degradation. The pretreatment mean gross tumour volume of these five brain metastases was 6.8cm3 (range: 3.3;14 cm3 ). The median delay of this complication was 8.8 months. Radionecrosis was suspected on two lesions in two patients in a mean delay of 9.5 months, confirmed by multimodal imaging in one case. The second patient died from other causes before we made the diagnosis. Both were treated by oral corticosteroids. 4. Discussion Each year, more than 150 patients are treated at our institution with stereotactic radiotherapy for one or more brain metastases. One-year local control rate was 88.3% in this series of patients selected to receive a 6 × 6 Gy scheme. It is consistent with those described previously in literature for hypofractionated stereotactic radiotherapy, with a comparable fractionation schedule: 76% at one year in a phase II study conducted by Ernstecken et al., and

Please cite this article in press as: Dumont Lecomte D, et al. Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions. Cancer Radiother (2019), https://doi.org/10.1016/j.canrad.2019.06.012

G Model

ARTICLE IN PRESS

CANRAD-3959; No. of Pages 7

D. Dumont Lecomte et al. / Cancer/Radiothérapie xxx (2019) xxx–xxx

4

Table 1 Study on hypofractionated stereotactic radiotherapy for challenging brain metastases: characteristics of the study population.

Age mean (± sd) median [range] Primary location Breast Lung Kidney Skin Head and neck Others WHO performance status 0–1 2–3 Number of brain metastases 1 2–3 4 RPA class 1 2 3 DS GPA class 0–1 1.5–2 2.5–3 3.5–4 Extracerebral metastases yes no Controlled extracerebral disease yes no Systemic treatment (◦ ongoing/*discontinued) Conventional chemotherapy Targeted therapy Immunotherapy Neurological symptoms before treatment yes Focal deficit Seizures Intracranial hypertension no

With haemorrhagic brain metastases (N = 21)

Without haemorrhagic brain metastases (N = 42)

All (N = 63)

P

63.62 years (± 9.87) 68 years [46–79]

63.83 years (± 12.07) 64 years [34–87]

63.76 years (± 11.31) 66 years [34–87]

0.94

2 (10%) 8 (38%) 4 (19%) 2 (1%) 2 (1%) 3 (14%)

5 (12%) 19 (45%) 5 (12%) 6 (14%) 1 (02%) 6 (14%)

7 (11%) 27 (43%) 9 (14%) 8 (13%) 3 (05%) 9 (14%)

17 (81%) 4 (19%)

34 (81%) 8 (19%)

51 (81%) 12 (19%)

8 (38%) 8 (38%) 5 (24%)

17 (4%) 17 (4%) 8 (19%)

25 (4%) 25 (4%) 13 (2%)

4 (19%) 17 (81%) 0 (0%)

14 (33) 24 (57%) 4 (1%)

18 (29%) 41 (65%) 4 (06%)

5 (31%) 6 (38%) 4 (25%) 1 (06%)

12 (33%) 12 (33%) 10 (28%) 2 (06%)

17 (33%) 18 (35%) 14 (27%) 3 (06%)

18 (86%) 3 (14%)

29 (69%) 13 (31%)

47 (75%) 16 (25%)

13 (72%) 5 (28%) 9 (◦ 4/*5) 2 (◦ 1/*1) 4 (◦ 0/*4) 3 (◦ 3/*0)

15 (52%) 14 (48%) 18 (◦ 12/*6) 9 (◦ 7/*2) 4 (◦ 0/*4) 5 (◦ 5/*0)

28 (6%) 19 (4%) 27 (◦ 17/*10) 11 (◦ 8%/*3) 8 (◦ 0/*8) 8 (◦ 8/*0)

9 (43%) 5 4 1 12 (57%)

22 (52%) 15 7 5 20 (48%)

31 (49%) 20 11 6 32 (51%)

0.8

1

0.91

0.14

1

0.22

0.16

0.57

0.48

WHO: World Health Organization; RPA: Recursive partitioning analysis; DS-GPA: diagnosis-specific graded prognostic assessment.

Table 2 Study on hypofractionated stereotactic radiotherapy for challenging brain metastases: characteristics of brain metastases in the study population.

Histology Adenocarcinoma Clear cell carcinoma Melanoma Ductal carcinoma Squamous cell carcinoma Others Gross tumour volume mean (± sd) median [range] Large brain metastases (> 15 cm3 ) yes no

Haemorrhagic brain metastases (N = 32)

Non haemorrhagic brain metastases (N = 60)

All (N = 92)

15 (47%) 5 (16%) 4 (12%) 2 (6%) 2 (6%) 4 (12%)

22 (37%) 10 (17%) 10 (17%) 10 (17%) 5 (8%) 3 (5%)

37 (40%) 15 (16%) 14 (15%) 12 (13%) 7 (8%) 7 (8%)

4.79 cm3 (±6.16 cm3 ) 2.15 cm3 [0.12–21.96]

8.42 cm3 (±9.32 cm3 ) 4.68 cm3 [0.28–42.84]

7.16 cm3 (±8.5 cm3 ) 4.07 cm3 [0.12–42.84]

3 (9%) 29 (91%)

9 (15%) 51 (85%)

12 (13%) 80 (87%)

P 0.55

0.027

0.53

69% at one year in Kim et al. [6,7]. Five patients were excluded because they were not evaluated by at least one MRI, but in this population only two died from neurological complications. Older data from randomized trials, which established the best tolerated dose of radiosurgery for the RTOG 90 05 trial reported local control rate at one year of about 70% [19]. The efficacy of radiosurgery alone versus radiosurgery with whole brain radiation therapy was shown by Aoyama et al. and the EORTC 22952–26001 trial with one-year

local control rates of 72.5% to about 70%, respectively [5,20]. In these studies, inclusion criteria were restrictive with respect to tumour size and number, and the question of intratumoral bleeding, either before or after treatment, was not addressed. We report relatively good local control for both lesions (compared to the literature cited above) presenting a pretreatment haemorrhagic signal and other brain metastases treated by a 6 × 6 Gy fractionation scheme. Besides, haemorrhagic brain

Please cite this article in press as: Dumont Lecomte D, et al. Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions. Cancer Radiother (2019), https://doi.org/10.1016/j.canrad.2019.06.012

G Model CANRAD-3959; No. of Pages 7

ARTICLE IN PRESS D. Dumont Lecomte et al. / Cancer/Radiothérapie xxx (2019) xxx–xxx

5

Fig. 1. Study on hypofractionated stereotactic radiotherapy for challenging brain metastases: local control and overall survival for the entire study population. a: cumulative incidence of local progression in our sample of 92 brain metastases treated with the 6 × 6 Gy regimen of hypofractionated stereotactic radiotherapy, (–) competing risk estimate, (–) 95% confidence interval; b: corresponding overall survival in the 63 patients, (–) Kaplan–Meier estimate, (–) 95% confidence interval. RT: radiotherapy.

Fig. 2. Study on hypofractionated stereotactic radiotherapy for challenging brain metastases: Local control and overall survival for haemorrhagic subgroups, and the rest of the population. a: cumulative incidence of local progression for haemorrhagic brain metastases and others, (–) non haemorrhagic (N = 60), (–) haemorrhagic (N = 32); b: corresponding overall survival for patients carrying at least one haemorrhagic brain metastasis and others, (–) non haemorrhagic (N = 42), (–) haemorrhagic (N = 21). RT: radiotherapy.

metastases were rather small in size in our population, which is not the most frequent situation, and were rarely symptomatic. As haemorrhagic metastases are often excluded from clinical trials, literature is limited in this category. Five studies specifically addressing outcomes of haemorrhagic brain metastases between 2003 and 2018 were identified in Medline database. These studies overall report a poorer prognosis in patients carrying haemorrhagic brain metastases, but treatment was each time radiosurgery and not hypofractionated stereotactic irradiation. In 2003, Suzuki et al. first described the occurrence of pre- and posttreatment haemorrhage, in a group of 54 patients presenting 131 brain metastases treated with a unique fraction of 20 to 25 Gy [21]. They reported a 7.4% rate of pretreatment haemorrhage but did not correlate their findings with local control. In 2007, Mathieu et al. report a hazard ratio of 2.5 of poorer local control rate for pretreatment haemorrhagic brain metastases, in a sample of 244 melanoma patients (37 patients had a haemorrhagic brain metastases before radiosurgery) [22]. Patients were treated by radiosurgery (45% of population previously receiving whole brain radiation therapy) using a median dose of 18 Gy. They hypothesized that their definition of target volumes was altered due to associated haemorrhage blurring the lesion margins, leading to an overvalued tumour size. Larger size may explain the poorer local control observed. However, in

multivariate analysis, large size of brain metastases (volume greater than 8 cm3 ) and haemorrhagic brain metastases were both significantly associated with poorer local control. In a series by Redmond et al. on 59 patients with 208 melanoma brain metastases, the preradiosurgery haemorrhagic brain metastases rate was 8.9% (N = 18) [23]. The local control rate was again influenced by previous bleeding. Moreover, results may have been skewed by the use of a previous whole brain radiation therapy for 40% of the population. Ghia et al. in 2014 specifically studied the influence of a pre- and postradiosurgery haemorrhagic signal in 110 patients presenting 358 lesions, whose 83 (23.4%) with an initial haemorrhagic signal [24]. The dose was adapted to tumour size based on a study by Shaw et al. [19]. Local control was also influenced by pretreatment haemorrhage, with local control rate at one year of 51.7% versus 64.9%, P = 0.03. In 2018, Bauer–Nilsen et al. reported outcomes of 936 melanoma brain metastases in 134 patients treated by stereotactic radiosurgery, using modalities of prescription leading to a mean margin dose of 18 to 20 Gy [25]. A poorer local control rate of 43% at 6 months in a subgroup of 58 haemorrhagic brain metastases versus 83% at 6 months in solid melanoma brain metastases was observed. To note, these two groups were significantly different in three aspects on multivariate analysis: the margin dose was lower in haemorrhagic brain metastases, the lesion size was

Please cite this article in press as: Dumont Lecomte D, et al. Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions. Cancer Radiother (2019), https://doi.org/10.1016/j.canrad.2019.06.012

G Model CANRAD-3959; No. of Pages 7 6

ARTICLE IN PRESS D. Dumont Lecomte et al. / Cancer/Radiothérapie xxx (2019) xxx–xxx

larger for haemorrhagic brain metastases (1.27cm3 versus 0.2cm3 in solid brain metastases, P < 0.001), and the prior use of chemotherapy: more frequent in the haemorrhagic subgroup. The originality of this study was that pretreatment MRI were reviewed to separate “clearly haemorrhagic metastases” from “melanin-containing metastases”, but that was not distinguish in efficacy results. Altogether, these results were interpreted as a sign that haemorrhagic brain metastases should be excluded and considered as inadequate candidates for stereotactic treatment. In our series, haemorrhagic complication occurred in 5.5% of the population. This rate is surprisingly low especially as brain metastases included 50% of lesions that were either already haemorrhagic or large. Suzuki et al. reported a 18.5% rate of haemorrhagic complication after radiosurgery [21]. They observed that large brain metastases were more likely to present this complication after the radiotherapy course, but it was not statistically significant. According to them, the important dose gradient produced by large doses per fraction may lead to central necrosis without obliteration of vascular structures, and lead to haemorrhage. Using smaller doses per fraction may prevent this complication, as our current study might suggest. In Mathieu et al. study, 38 patients had haemorrhagic complications after radiosurgery (18.4% of 206 patients) [22]. Multivariate analysis established that this complication rate was only associated with tumour size. In Redmond et al. study, haemorrhagic complications occurred in nine lesions (4.3%) and was not significantly associated with local control rate [23]. In Ghia et al. study, 78 brain metastases (20.4%) had post-treatment haemorrhagic complications, and worse one-year local control rate: 32.7% versus 67.8%, P < 0.001 [24]. With only five patients in our study, the impact on local control was not evaluated. The median delay was quite long, with 8.8 months in our patients, as Suzuki et al. reported half the occurrence within a month (range: 1 day to 4 months) and Mathieu et al. a median delay of 1.8 months [21,22]. Spontaneous bleedings are also possible at any times, and a long onset of appearance after treatment may result from the natural history of the disease rather than a toxic effect. No grade 3 toxicity was observed in the study by Bauer–Nilsen et al. and postprocedure haemorrhagic complications were not reported [25]. The 6 × 6 Gy hypofractionated stereotactic radiotherapy regimen was well tolerated in our patient sample, with only one case of severe and finally regressive acute toxicity after symptomatic treatment. Thirty-one symptomatic patients were under oral steroids before starting hypofractionated stereotactic radiotherapy. This treatment was established in almost all patients during the hypofractionated stereotactic radiotherapy course as we usually do. Five patients did not receive any steroids at all, as they were taking concomitant immunotherapy and carrying low risk brain metastases (quite small size tumour without brain oedema). High blood pressure and anticoagulant therapy are two factors that may interfere with the risk of cerebral haemorrhage, they should be taking in account in further studies for more accurate inter comparisons. Radionecrosis rate was low in our series, compared to previously published results. More follow up is needed to conclude clearly on this point, since the occurrence of radionecrosis may be later. The volume of brain receiving 26 Gy (i.e. V26 Gy) with the 6 × 6 Gy regimen corresponds to the volume of brain receiving 14 Gy (V14 Gy) described by Inoue et al. for a 3 × 10 Gy regimen [26]. The V26 was greater than 7 cm3 in the two patients who experienced radionecrosis. With comparable fractionation schedule, the incidence of radionecrosis is comparable to the results from a series by Jimenez et al. on five patients out of 156 [27]. Of note however, their definition of radionecrosis was not exactly the same (more than 6 months use of systemic corticoids or surgical evaluation). Survival at 6 months and one year were consistent for patients presenting at least one haemorrhagic brain metastases and others. In our study, performance status 2 or above and age 66 years or

older were the only significant prognostic factors associated with a worse survival. Neither brain metastases number, histology subtypes, controlled extracerebral disease, or prognostic scores were associated with overall survival. Intriguingly, there were similar outcomes of patients with or without cerebral salvage treatment. Our study results are limited in several ways as in any retrospective chart review study. Improvements in imaging, especially MRI techniques, reveal that microbleeding is probably underestimated in brain metastases [28], and may be more important in prior studies cited, where modality to detect haemorrhagic signal is not always clearly described. Brain metastases with small haemorrhagic signal have been probably treated in several studies. It is likely that there are many different types of haemorrhagic brain metastases and taken aside those who need to be surgically resected, we need to investigate which haemorrhagic lesion can benefit or not from hypofractionated stereotactic radiotherapy. 5. Conclusion The 6 × 6 Gy fractionation scheme seems to be a safe and effective regimen in the treatment of brain metastases at risk for complications. Haemorrhagic brain metastases are frequent, especially in melanoma and renal clear cell carcinoma. The treatment of these lesions seems quite safe with this scheme, and local control rate at one year was excellent. We suggest that further studies should be made to confirm these results in a prospective way. With survival improving in these pathologies, we need to offer the same appropriate treatment for patient presenting haemorrhagic brain metastases than the others, especially to preserve neurocognitive functions. A French grant has been obtained recently to conduct prospectively this evaluation in two centres including ours. Contribution D Dumont Lecomte: conceptualization, data curation, investigation, methodology, visualization, writing original draft, review and editing; J Lequesne: data curation, formal analysis, methodology, software, visualization; J Geffrelot, V Barraux, C Loiseau: investigation, resources; P Lesueur: investigation, resources, writing review and editing; J Lacroix: investigation, resources, supervision; A. Leconte: data curation, methodology, project administration; É Émery: supervision, writing review and editing; J Thariat: conceptualization, data curation, methodology, supervision, validation, visualization, writing review and editing; D Stefan: conceptualization, data curation, methodology, resources, supervision, validation, visualization, writing review and editing. Disclosure of interest The authors declare that they have no competing interest. References [1] Taillibert S, Le Rhun É. Epidemiology of brain metastases. Cancer Radiother 2015;19:3–9. [2] Gaspar LE, Scott C, Murray K, Curran W. Validation of the RTOG recursive partitioning analysis (RPA) classification for brain metastases. Int J Radiat Oncol Biol Phys 2000;47:1001–6. [3] Sperduto PW, Kased N, Roberge D, Xu Z, Shanley R, Luo X, et al. Summary report on the graded prognostic assessment: an accurate and facile diagnosisspecific tool to estimate survival for patients with brain metastases. J Clin Oncol 2012;30:419–25. [4] Yamamoto M, Serizawa T, Shuto T, Akabane A, Higuchi Y, Kawagishi J, et al. Stereotactic radiosurgery for patients with multiple brain metastases (JLGK0901): a multi-institutional prospective observational study. Lancet Oncol 2014;15:387–95. [5] Aoyama H, Shirato H, Tago M, Nakagawa K, Toyoda T, Hatano K, et al. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic

Please cite this article in press as: Dumont Lecomte D, et al. Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions. Cancer Radiother (2019), https://doi.org/10.1016/j.canrad.2019.06.012

G Model CANRAD-3959; No. of Pages 7

ARTICLE IN PRESS D. Dumont Lecomte et al. / Cancer/Radiothérapie xxx (2019) xxx–xxx

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA 2006;295:2483–91. Ernst-Stecken A, Ganslandt O, Lambrecht U, Sauer R, Grabenbauer G, Phase II. trial of hypofractionated stereotactic radiotherapy for brain metastases: results and toxicity. Radiother Oncol 2006;81:18–24. Kim Y-J, Cho KH, Kim J-Y, Lim YK, Min HS, Lee SH, et al. Single-dose versus fractionated stereotactic radiotherapy for brain metastases. Int J Radiat Oncol 2011;81:483–9. Murai T, Ogino H, Manabe Y, Iwabuchi M, Okumura T, Matsushita Y, et al. Fractionated stereotactic radiotherapy using CyberKnife for the treatment of large brain metastases: a dose escalation study. Clin Oncol R Coll Radiol 2014;26:151–8. Brown PD, Jaeckle K, Ballman KV, Farace E, Cerhan JH, Anderson SK, et al. Effect of radiosurgery alone vs radiosurgery with whole brain radiation therapy on cognitive function in patients with 1 to 3 brain metastases: a randomized clinical trial. JAMA 2016;316:401–9. Jiang X, Xiao J, Zhang Y, Xu Y, Li X, Chen X, et al. Hypofractionated stereotactic radiotherapy for brain metastases larger than three centimeters. Radiat Oncol 2012;7:36. Feuvret L, Vinchon S, Martin V, Lamproglou I, Halley A, Calugaru V, et al. Stereotactic radiotherapy for large solitary brain metastases. Cancer Radiother 2014;18:97–106. Sinclair G, Bartek J, Martin H, Barsoum P, Dodoo E. Adaptive hypofractionated gamma knife radiosurgery for a large brainstem metastasis. Surg Neurol Int 2016;7:S130–8. Soffietti R, Abacioglu U, Baumert B, Combs SE, Kinhult S, Kros JM, et al. Diagnosis and treatment of brain metastases from solid tumors: guidelines from the European Association of Neuro-Oncology (EANO). Neuro-Oncol 2017;19:162–74. Hall EJ, Brenner DJ. The radiobiology of radiosurgery: rationale for different treatment regimes for AVMs and malignancies. Int J Radiat Oncol Biol Phys 1993;25:381–5. Baliga S, Garg MK, Fox J, Kalnicki S, Lasala PA, Welch MR, et al. Fractionated stereotactic radiation therapy for brain metastases: a systematic review with tumour control probability modelling. Br J Radiol 2017;90:20160666. Zhang W, Ma X-X, Ji Y-M, Kang X-S, Li C-F. Haemorrhage detection in brain metastases of lung cancer patients using magnetic resonance imaging. J Int Med Res 2009;37:1139–44. Noël G, Simon JM, Valery CA, Cornu P, Boisserie G, Hasboun D, et al. Radiosurgery for brain metastasis: impact of CTV on local control. Radiother Oncol 2003;68:15–21.

7

[18] Wiggenraad R, Kanter AV, Mast M, Molenaar R, Kal HB, Nijeholt GL, et al. Local progression and pseudo progression after single fraction or fractionated stereotactic radiotherapy for large brain metastases. Strahlenther Onkol 2012;188:696–701. [19] Shaw E, Scott C, Souhami L, Dinapoli R, Kline R, Loeffler J, et al. Single dose radiosurgical treatment of recurrent previously irradiated primary brain tumors and brain metastases: final report of RTOG protocol 90-05. Int J Radiat Oncol Biol Phys 2000;47:291–8. [20] Kocher M, Soffietti R, Abacioglu U, Villà S, Fauchon F, Baumert BG, et al. 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 2011;29:134–41. [21] Suzuki H, Toyoda S, Muramatsu M, Shimizu T, Kojima T, Taki W. Spontaneous haemorrhage into metastatic brain tumours after stereotactic radiosurgery using a linear accelerator. J Neurol Neurosurg Psychiatr 2003;74:908–12. [22] Mathieu D, Kondziolka D, Cooper PB, Flickinger JC, Niranjan A, Agarwala S, et al. Gamma knife radiosurgery in the management of malignant melanoma brain metastases. Neurosurgery 2007;60:471–81 [discussion 481-2]. [23] Redmond AJ, Diluna ML, Hebert R, Moliterno JA, Desai R, Knisely JPS, et al. Gamma Knife surgery for the treatment of melanoma metastases: the effect of intratumoral hemorrhage on survival. J Neurosurg 2008;109:99–105. [24] Ghia AJ, Tward JD, Anker CJ, Boucher KM, Jensen RL, Shrieve DC. Radiosurgery for melanoma brain metastases: the impact of hemorrhage on local control. J Radiosurgery SBRT 2014;3:43–50. [25] Bauer-Nilsen K, Trifiletti DM, Chatrath A, Ruiz-Garcia H, Marchan E, Peterson J, et al. Stereotactic radiosurgery for brain metastases from malignant melanoma and the impact of hemorrhagic metastases. J Neurooncol 2018;140:83–8. [26] Inoue HK, Seto K-I, Nozaki A, Torikai K, Suzuki Y, Saitoh J-I, et al. Threefraction CyberKnife radiotherapy for brain metastases in critical areas: referring to the risk evaluating radiation necrosis and the surrounding brain volumes circumscribed with a single dose equivalence of 14 Gy (V14). J Radiat Res 2013;54:727–35. [27] Jimenez RB, Alexander BM, Mahadevan A, Niemierko A, Rajakesari S, Arvold ND, et al. The impact of different stereotactic radiation therapy regimens for brain metastases on local control and toxicity. Adv Radiat Oncol 2017;2:391–7. [28] Franceschi AM, Moschos SJ, Anders CK, Glaubiger S, Collichio FA, Lee CB, et al. Utility of susceptibility weighted imaging (SWI) in the detection of brain hemorrhagic metastases from breast cancer and melanoma. J Comput Assist Tomogr 2016;40:803–5.

Please cite this article in press as: Dumont Lecomte D, et al. Hypofractionated stereotactic radiotherapy for challenging brain metastases using 36 Gy in six fractions. Cancer Radiother (2019), https://doi.org/10.1016/j.canrad.2019.06.012