Bone invasiveness is associated with prognosis in clivus chordomas

Journal of Clinical Neuroscience 27 (2016) 147–152

Contents lists available at ScienceDirect

Journal of Clinical Neuroscience journal homepage: www.elsevier.com/locate/jocn

Clinical Study

Bone invasiveness is associated with prognosis in clivus chordomas Ke Wang, Liang Wang, Zhen Wu, Kaibing Tian, Cong Ren, Guijun Jia, Liwei Zhang, Junting Zhang ⇑ Department of Neurosurgery, Beijing Tian Tan Hospital, Capital Medical University, 6 Tiantan Xili, Dongcheng District, Beijing 100050, PR China China National Clinical Research Center for Neurological Diseases, Beijing 100050, PR China

a r t i c l e

i n f o

Article history: Received 10 April 2015 Accepted 30 May 2015

Keywords: Clivus chordoma Growth pattern Prognosis

a b s t r a c t Clivus chordomas present a great challenge for neurosurgeons, and the prognosis is poor. To investigate bone invasiveness characteristics in regard to the prognosis of clivus chordomas, a retrospective study of 19 patients with primary clivus chordoma was performed. Clinical data, MRI, CT scans and follow-up data were examined, and the bone invasiveness of the chordoma was classified into two growth patterns (endophytic and exophytic) which were analyzed with regard to prognosis. The overall survival rate was 78.9% with radical surgery and adjuvant radiation, with a mean follow-up of 44.5 months. There were 12 patients in the endophytic group and seven in the exophytic group, and the exophytic group exhibited a higher recurrence rate than the endophytic group (p = 0.006). Chordomas with an exophytic growth pattern were more likely to recur than those with an endophytic growth pattern, and the surgical approach can be tailored according to each growth pattern. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction Chordoma is a rare tumor derived from the remnants of the notochord with an estimated incidence of 0.08 per million in Western nations [1] and 0.04 per million in Taiwan Han Chinese populations [2]. They usually occur in the axis, and 49.4% occur in the clivus regions [3]. The treatment for clivus chordomas is challenging, and total removal is ideal [4–8]. However, the slowgrowing, yet infiltrative, characteristics of this relatively low malignancy tumor makes complete removal difficult, particularly because of the surrounding vital neurovascular structures, which contributes to recurrence and poor prognosis [8,9]. One systematic review found 5 year and 10 year survival rates were 70% and 63%, respectively [9]. Modern technologies and improvements in neurosurgery have resulted in several well-documented approaches for treatment, the goals of which are to remove as much tumor as possible, improve quality of life and prolong survival [6,8]. Anterior approaches, lateral approaches, and multi-staged combined approaches are used extensively according to the site of the tumor and its relationship to the clivus [4–8,10–13]. Our previous report [8] demonstrated that radical surgery contributed to a longer survival. However, a case series by Sen et al. [7] reported a survival rate of only 58% with radical resection, and radical resection was

⇑ Corresponding author. Tel.: +86 10 6709 8431; fax: +86 10 6709 1377. E-mail address: [email protected] (J. Zhang). http://dx.doi.org/10.1016/j.jocn.2015.05.073 0967-5868/Ó 2015 Elsevier Ltd. All rights reserved.

dependent on preoperative tumor volume and the number of anatomical areas involved with tumor. Our previous report [8] classified these tumors into four states according to their dura mater infiltration. However, dura mater infiltration was not an independent risk factor for prognosis. This classification was insufficient because it did not consider the relationship of the tumor to bone (clivus). Bone invasiveness makes complete surgical removal difficult to achieve, which contributes to tumor recurrence. Rarely, case reports demonstrate that chordoma may arise in the intradural space [14,15]. MRI and CT scans reveal the various relationships between these tumors and nearby bones. Our recent proteomics study showed that the growth pattern of clivus chordomas (endophytic and exophytic types) may be differentiated by transforming growth factor b (TGFb) and phosphatase and tensin homolog deleted on chromosome 10 (PTEN) [16]. The clinical importance of these two growth patterns is not well documented. Therefore, we retrospectively reviewed our own series of primary clivus chordomas to investigate this characteristic. 2. Materials and methods 2.1. Patient population The senior author treated 19 patients with primary clivus chordoma in the Department of Neurosurgery, Tian Tan Hospital, Capital Medical University, China from February 2004 to June 2005. All cases were retrospectively studied using detailed data.

148

K. Wang et al. / Journal of Clinical Neuroscience 27 (2016) 147–152

Cases without pathological confirmation were excluded. Patient follow-ups were performed via postsurgical office visits and telephone interviews with patients who were unable to visit in person. All patient data were compiled from hospital and office records, imaging studies, and the patients’ own records. All patients voluntarily participated in the study under the protection of the Hospital Ethics Committee. 2.2. Classification of growth pattern The growth pattern of the clivus chordoma was identified according to the bone invasiveness of the tumor on preoperative images, including plain and enhanced head MRI, thin layer skull base CT scanning and three-dimensional reconstruction, as described in a previous report [16]. The chordomas were

classified into two subtypes: endophytic and exophytic, as shown in Figure 1 and 2. The following methods were used: the maximum diameter at eyeball level on T2-weighted axial MRI was defined as the baseline level, and the area of the bilateral carotid cavernous lateral walls connected to the bilateral petrous apex at the baseline was used as the standard region. A tumor was identified as endophytic if at least 50% of the tumor, which can invade the bone in all directions, was located within the standard region at the baseline level or if the clivus bone was transformed into a ‘‘bubble” or ‘‘dumbbell” (Fig. 1A, 2A,C). Tumors were identified as exophytic if at least 50% of the tumor was located outside the standard region with limited bone invasiveness at baseline levels; these tumors may show a ‘‘pedicle-like” attachment to the clivus into intracranial areas on MRI and CT scans (Fig. 1B, 2B,D).

Fig. 1. Growth pattern in clivus chordomas. (A) Axial T2-weighted and (B) sagittal T1-weighted with enhancement MRI of an endophytic clivus chordoma. (C) Axial T2weighted and (D) sagittal T1-weighted with enhancement MRI of an exophytic clivus chordoma.

K. Wang et al. / Journal of Clinical Neuroscience 27 (2016) 147–152

149

Fig. 2. Schematic based on sagittal and axial imaging of the classification of the endophytic and exophytic growth patterns. (A, C) A tumor was identified as endophytic if at least 50% of the tumor, which can invade the bone in all directions, was located within the standard region (red rectangle) at the baseline level or if the clivus bone was transformed into a ‘‘bubble” or ‘‘dumbbell”. (B, D) Tumors were identified as exophytic if at least 50% of the tumor was located outside of the standard region with limited bone invasiveness at baseline levels; these tumors may show a ‘‘pedicle-like” attachment to the clivus into intracranial areas on MRI and CT scans.

2.3. Statistical analysis Clinical and follow-up data were collected using database software (EpiData version 3.02; EpiData, Odense, Denmark). Kaplan–Meier survival analysis and chi-square tests were performed using the Statistical Package for the Social Sciences version 13.0 (IBM, Armonk, NY, USA). 3. Results Nineteen primary clivus chordomas were pathologically confirmed, and all tumors were of the classic type. Table 1 shows the details of every patient. Surgeries were performed with the goal of removing as much of the tumor as safely possible. Four patients died, three from excessive brain edema (one patient died during first surgery, and the other two patients died during surgery to remove a rapid recurrence) and one patient died from rapid recurrence after the first surgery (he declined further surgery). Tumor volumes ranged from 5 to 180 ml3 (D1  D2  D3/2), with a mean volume of 70.5 ml3, and most tumors were subtotally removed. Adjuvant radiation was performed in eight patients (8/18, 44%) 3 months after surgery: in four the Gamma Knife (Elekta AB, Stockholm, Sweden) was used; three received traditional radiation treatment; and one patient received stereotactic radiosurgery. Follow-up periods ranged from 2 to 122 months (mean 44.5 months). Most patients had a Karnofsky Performance Status (KPS) score greater than 80 at the last follow-up, although seven patients had a follow-up period of less than 12 months (including the four patients who died). Recurrence was observed on MRI in five patients, within a median follow-up of 45 months (range 1–122 months). In three patients, recurrence occurred within 6 months after surgery, and

in the other two patients recurrence occurred at 36 and 47 months after surgery. Six patients with follow-up periods >96 months recovered (KPS P 90) and returned to work or study. All 19 primary clivus chordomas fell into the two bone invasiveness groups according to our classification: the endophytic group (12 patients) and exophytic group (seven patients). In all cases, patients underwent surgery, which was most commonly subtotal removal. In only three cases did patients receive partial removal (one in the endophytic group, and two in the exophytic group). Table 2 shows that surgical resection was similar in both groups (P = 0.243). There was no significant difference in recurrence or prognosis rates. Kaplan–Meier univariate analysis revealed that the exophytic group was more likely to recur (P = 0.006, Fig. 3). 4. Discussion Clivus chordomas are relatively low malignancy tumors that are slow-growing and infiltrative [17]. These tumors grow slowly in most cases, but the overall survival is poor. The Surveillance, Epidemiology, and End Results database showed that chordoma patients had a median survival of 6.29 years and that the 5, 10, and 20 year survival rates drop precipitously to 67.6%, 39.9%, and 13.1%, respectively [1]. Clivus chordomas that are surrounded by vital neurovascular structures are difficult to remove in most cases, and surgery is associated with high rates of complications and mortality, including cranial nerve palsy, brain stem infarction and internal carotid artery injury. Our previous report on the patients included in this paper showed that the 5 and 10 year survival rates were 67.6% and 59.5%, respectively, and the recurrence rates at 5 and 10 years were 52.9% and 88.3%, respectively [8]. The present series consisting of all primary classic-type clivus chordomas with a mean follow-up of 45 months, revealed a recurrence of 26.3% (five patients) and an overall survival rate of 78.9% (15/19),

150

K. Wang et al. / Journal of Clinical Neuroscience 27 (2016) 147–152

Table 1 Demographic and clinical details of 19 patients with clivus chordomas Surgical approach

Extent of resection

Adjuvant radiation

Recurrence

Treatment for recurrence

Follow-up (months)

State at last follow-up

30

FL

Subtotal

Traditional

36 months

r-knife

110

Exophytic

5

MF

Partial

None

No

NA

20

43

Exophytic

24

PS

Subtotal

Traditional

No

NA

38

Female

27

Exophytic

75

MF

Subtotal

Traditional

No

NA

2

5 6

Male Male

53 54

Exophytic Endophytic

24 16

P PS

Partial Subtotal

None None

6 months 47 months

surgery surgery

6 104

7

Male

25

Endophytic

126

PS

Subtotal

SRS

No

NA

105

8 9

Female Male

19 37

Exophytic Exophytic

75 75

MF PS

Total Subtotal

SRS None

6 months No

surgery NA

10

Male

42

Endophytic

105

MF

Subtotal

None

No

NA

61

11

Male

37

Endophytic

105

PS

Subtotal

None

No

NA

109

12 13

Female Female

35 11

Endophytic Endophytic

25 180

FL PS

Subtotal Subtotal

None SRS

No No

NA NA

6 112

14

Female

9

Endophytic

40

FL

Partial

Gamma Knifea

No

NA

122

15

Female

22

Endophytic

125

FL

Subtotal

SRS

No

NA

12

16 17 18 19

Female Male Male Male

21 38 27 28

Endophytic Exophytic Endophytic Endophytic

168 50 30 15

PS PS PS MF

Subtotal Subtotal Subtotal Subtotal

None None None None

No 6 months NA No

NA Tradition NA NA

12 6 1 12

Alive without tumor Alive without tumor Alive without tumor Alive without tumor Dead Alive with tumor Alive with tumor Dead Alive without tumor Alive with tumor Alive with tumor Dead Alive with tumor Alive without tumor Alive without tumor Missing Missing Dead Alive without tumor

Patient

Sex

Age (years)

Group

Tumor volume (ml)

1

Male

32

Endophytic

2

Female

36

3

Female

4

6 2

KPS at follow-up 90

80

80

80

0 90 100 0 80

90 90 0 100 90

80

80 80 0 80

a Elekta AB, Stockholm, Sweden. Follow-up times were different as some patients could not be contacted in April 2014. In such cases the latest follow-up documents were used. FL = far lateral approach, KPS = Karnofsky Performance Status, MF = middle fossa approach, NA = not applicable, P = pterional approach, PS = pre-sigmoid approach, SRS = stereotactic radiosurgery.

Table 2 Characteristics of endophytic and exophytic clivus chordomas

Sex Extent of resection Adjuvant radiation Recurrence Stage * #

Male Female Radical* Partial Yes No Yes No Dura mater integrity Dura mater breach

Endophytic group

Exophytic group

P value#

7 5 11 1 5 7 2 10 6 6

3 4 5 2 3 4 3 4 2 5

0.515 0.243 0.960 0.211 0.361

Total and subtotal. Chi-square test with Fisher’s exact test.

similar to our previous report and reports in the literature [5,7–9]. The treatment strategy was radical removal in the early period, as advocated by many previous reports [6,7,9]. In the present series total and subtotal removal to remove as much of each tumor as possible using a pre-sigmoid or middle fossa approach was the treatment strategy for most patients. Previous studies have

reported higher or similar radical surgical rates for clivus chordoma using ventral approaches with limited complications compared to traditional transcranial approaches [10,13]. However, transcranial and ventral approaches are aimed at radical tumor resection, and both approaches contribute to improved outcomes for clivus chordomas [11,12]. Neurosurgeons should be cautious

K. Wang et al. / Journal of Clinical Neuroscience 27 (2016) 147–152

Fig. 3. Kaplan–Meier univariate analysis of survival of endophytic and exophytic clivus chordomas. Cum = cumulative.

in treatment decision-making in chordomas to avoid surgical complications which affect long-term quality of life, and the complicated anatomical relationship to the skull base should always be considered. The infiltrative growth pattern of clivus chordomas may occur slowly, or it may present as rapid growth in some patients [15,18,19]. Chordomas are prone to rapid recurrence when Ki-67 levels are higher than 5% [15], with up-regulated N-cadherin and down-regulated E-cadherin [18,19]. This phenomenon is especially obvious when the tumor infiltrates the dura mater into the subdural spaces. Thus, the various growth patterns may influence prognosis. Tumors in the present series were classified according to their bone invasiveness into two groups: the exophytic group and the endophytic group. Tumors in the endophytic group primarily infiltrated the bone. These tumors are difficult to remove radically because there are no boundaries between tumor and bone. Sometimes, the tumor grows in separate ‘‘chambers” or pockets of the bone, which makes it extremely difficult to determine the degree of resection. On the other hand, exophytic tumors exhibit limited bone infiltration. This type of chordoma can grow into the brain stem and surround neurovascular structures, which creates high surgical risk. However, the exophytic type of chordoma may be radically removed from the involved bones, with special attention paid to the brain stem and surrounding neurovasculature. However, results based on the present data demonstrate that patients in the exophytic group had a relatively poor prognosis in both the perioperative period and long-term. The reason for the poorer prognosis may lie in the incomplete resection and relatively high growth characteristic of these tumors. Firstly, the role of extent of removal as an independent prognostic factor has been well illustrated in the literature, and incomplete resection results in recurrence and a poor prognosis [3–5,8,9,20]. Secondly, the intrinsic characteristics of the exophytic chordomas may result in uncontrollable tumor progression. A previous study demonstrated that TGFb and PTEN can contribute to differences in the growth patterns of clivus chordomas [16]. Other studies have reported a rapid progression after the initial removal, and numerous molecules may be involved, including Ki-67, cadherins, mTOR, and c-MET [15,18,19,21,22]. In contrast, the endophytic growth pattern may grow at a relatively slow rate, indicating that these tumors are of a more indolent type [15]. The extent of resection was similar in both groups in the present series, but it is likely that exophytic clivus chordomas may be related to a poor prognosis. However, further studies are required to investigate this hypothesis.

151

One limitation of the present study is that all tumors were transcranially treated, with no ventral approaches used. Currently, clivus chordomas may be treated using various surgical approaches depending on their growth pattern. Notably, ventral approaches have progressed rapidly in recent years, and high-definition endoscopic tools can help expose the entire central cranial base from the crista galli to the odontoid process, as well as laterally to the jugular vein and foramen magnum [23]. The present growth pattern classification provides a rationale that may aid in decisionmaking. Endophytic growth clivus chordomas can be surgically treated using ventral approaches with the help of an endoscope [12,13]. Treatment of exophytic clivus chordomas using ventral approaches may be complicated and its efficacy debatable, considering the lateral extension and brain stem involvement. Lateral, combined, or staged approaches may be used based on the experience of the neurosurgeon, the aim of the surgery, and at the wishes of the patient [7,8,12,23]. The patients enrolled in the current study were chosen to analyze the first treatment strategies applied early in our senior surgeons’ career. The present data and our previous report [24] show that patients in the endophytic group have a longer overall survival rate with relatively good KPS if treated with radical surgery with protection of the dura mater, partially due to the slow-growing characteristic of these tumors. The present classification of endophytic and exophytic growth patterns is different from that used for intradural clivus chrodomas. As an embryonic structure, the human notochord undergoes significant regression by birth and complete involution by the end of the first decade of life. However, remnants of the notochord can be found in approximately 20% of adult spines [25]. Benign notochordal cell tumors have an anatomic distribution similar to classic chordomas, and these tumors may act as a precursor to chordoma [26]. It is believed that the chordoma derives from the notochord, with the tumor infiltrating the surrounding bones [15], however, the origin of some tumors has been reported to be intra-dural [14,15]. Exophytic clivus chordomas in the present study were different from intradural clivus chordomas. The exophytic clivus chordomas were identified based on an imaging classification of bone invasion. Endophytic and exophytic chordomas are of extradural origin [1,2]. In theory, a chordoma with an exophytic growth pattern may occur outside of the clivus, that is, from the clivus to the nasopharynx area [10]. This type of chordoma may be treated by ear, nose and throat surgeons, but there were no tumors of this type in the present study. Adjuvant radiotherapy is currently under much debate. Some authors argue the radiation resistance of clivus chordomas, and there is little evidence supporting radiotherapy after complete removal of the tumor [3,9,27]. However, radiotherapy may be indicated in patients with incomplete removal [28,29], particularly proton beam therapy [28] or charged particle radiation such as carbon ion radiation [30,31]. Patients with clivus chordomas in the present study were radically treated with the greatest extent of resection possible, with an observation period of 3 months or longer after surgery. Radiation therapy was recommended when a recurrence or residual tumor was present on imaging. This strategy was implemented to avoid unnecessary radiotherapy of the surgical area and radiation complications. A 10 year meta-analysis of observational studies revealed no survival differences between radiation therapy techniques and similar rates of progression-free survival and overall survival at 5 years between proton-beam, carbon ion, and modern fractionated photon radiation therapies [9]. The present data are consistent with these reports. We found that recent advances in radiotherapy can contribute to long-term tumor control for clivus chordomas. However, many such radiotherapy techniques are too expensive [29,30]. Overall, the treatment

152

K. Wang et al. / Journal of Clinical Neuroscience 27 (2016) 147–152

strategy of radical surgery with adjuvant radiotherapy in residual or recurrent tumors is advocated [8]. 5. Conclusion The present data demonstrated different growth patterns in clivus chordomas, and the exophytic growth pattern seems to predict poor prognosis. Surgical approaches may be selected depending upon growth pattern and dura mater infiltration. However, these findings require further investigation in a large series. Conflicts of Interest/Disclosures The authors declare that they have no financial or other conflicts of interest in relation to this research and its publication. Acknowledgments We are grateful to the study subjects for participating in this research and to everyone in Beijing Tian Tan Hospital who contributed to the study. This work was supported in part by the Natural Science Foundation of China (grant no. 81101910) and the Natural Science Foundation of Beijing, China (grant no. 7142052). References [1] McMaster ML, Gildstein AM, Bromley CM, et al. Chordoma: incidence and survival patterns in the United States, 1973–1995. Cancer Causes Control 2001;12:1–11. [2] Hung GY, Horng JL, Yen HJ, et al. Incidence patterns of primary bone cancer in Taiwan (2003–2010): a population-based study. Ann Surg Oncol 2014;21:2490–8. [3] Lee J, Bhatia NN, Hoang BH, et al. Analysis of prognostic factors for patients with chordoma with use of the California Cancer Registry. J Bone Joint Surg Am 2012;94:356–63. [4] Choi D, Melcher R, Harms J, et al. Outcome of 132 operations in 97 patients with chordomas of the craniocervical junction and upper cervical spine. Neurosurgery 2010;66:59–65 [discussion 65]. [5] Di Maio S, Rostomily R, Sekhar LN. Current surgical outcomes for cranial base chordomas: cohort study of 95 patients. Neurosurgery 2012;70:1355–60 [discussion 1360]. [6] Eid AS, Chang UK, Lee SY, et al. The treatment outcome depending on the extent of resection in skull base and spinal chordomas. Acta Neurochir (Wien) 2011;153:509–16. [7] Sen C, Triana AI, Berglind N, et al. Clival chordomas: clinical management, results, and complications in 71 patients. J Neurosurg 2010;113:1059–71. [8] Wu Z, Zhang J, Zhang L, et al. Prognostic factors for long-term outcome of patients with surgical resection of skull base chordomas-106 cases review in one institution. Neurosurg Rev 2010;33:451–6. [9] Di Maio S, Temkin N, Ramanathan D, et al. Current comprehensive management of cranial base chordomas: 10-year meta-analysis of observational studies. J Neurosurg 2011;115:1094–105.

[10] Carrabba G, Dehdashti AR, Gentili F. Surgery for clival lesions: open resection versus the expanded endoscopic endonasal approach. Neurosurg Focus 2008;25:E7. [11] Chau AM, Lazzaro A, Mobbs RJ, et al. Combined endoscopic endonasal and posterior cervical approach to a clival chordoma. J Clin Neurosci 2010;17:1463–5. [12] Komotar RJ, Starke RM, Raper DM, et al. The endoscope-assisted ventral approach compared with open microscope-assisted surgery for clival chordomas. World Neurosurg 2011;76:318–27 [discussion 259–62]. [13] Koutourousiou M, Gardner PA, Tormenti MJ, et al. Endoscopic endonasal approach for resection of cranial base chordomas: outcomes and learning curve. Neurosurgery 2012;71:614–24 [discussion 624–5]. [14] Ito E, Saito K, Nagatani T, et al. Intradural cranial chordoma. World Neurosurg 2010;73:194–7 [discussion e31]. [15] Wang L, Wu Z, Tian K, et al. Clinical and pathological features of intradural retroclival chordoma. World Neurosurg 2014;82:791–8. [16] Wu Z, Wang L, Guo Z, et al. Experimental study on differences in clivus chordoma bone invasion: an iTRAQ-based quantitative proteomic analysis. PLoS ONE 2015;10:e0119523. [17] Di Maio S, Kong E, Yip S, et al. Converging paths to progress for skull base chordoma: review of current therapy and future molecular targets. Surg Neurol Int 2013;4:72. [18] Karamchandani J, Wu MY, Das S, et al. Highly proliferative sellar chordoma with unusually rapid recurrence. Neuropathology 2013;33:424–30. [19] Triana A, Sen C, Wolfe D, et al. Cadherins and catenins in clival chordomas: correlation of expression with tumor aggressiveness. Am J Surg Pathol 2005;29:1422–34. [20] Ito E, Saito K, Okada T, et al. Long-term control of clival chordoma with initial aggressive surgical resection and gamma knife radiosurgery for recurrence. Acta Neurochir (Wien) 2010;152:57–67 [discussion 67]. [21] Akhavan-Sigari R, Gaab MR, Rohde V, et al. Expression of PDGFR-a, EGFR and cMET in spinal chordoma: a series of 52 patients. Anticancer Res 2014;34:623–30. [22] Chen K, Mo J, Zhou M, et al. Expression of PTEN and mTOR in sacral chordoma and association with poor prognosis. Med Oncol 2014;31:886. [23] Chibbaro S, Cornelius JF, Froelich S, et al. Endoscopic endonasal approach in the management of skull base chordomas–clinical experience on a large series, technique, outcome, and pitfalls. Neurosurg Rev 2014;37:217–24 [discussion 224–5]. [24] Wang K, Wang L, Tian K, et al. Surgical resection of upper-middle clivus chordomas via a modified anterior transpetrous approach. Clin Neurol Neurosurg 2015;130:20–5. [25] Yamaguchi T, Suzuki S, Ishiiwa H, et al. Intraosseous benign notochordal cell tumours: overlooked precursors of classic chordomas? Histopathology 2004;44:597–602. [26] Deshpande V, Nielsen GP, Rosenthal DI, et al. Intraosseous benign notochord cell tumors (BNCT): further evidence supporting a relationship to chordoma. Am J Surg Pathol 2007;31:1573–7. [27] Jian BJ, Bloch OG, Yang I, et al. A comprehensive analysis of intracranial chordoma and survival: a systematic review. Br J Neurosurg 2011;25:446–53. [28] Jian BJ, Bloch OG, Yang I, et al. Adjuvant radiation therapy and chondroid chordoma subtype are associated with a lower tumor recurrence rate of cranial chordoma. J Neurooncol 2010;98:101–8. [29] Yasuda M, Bresson D, Chibbaro S, et al. Chordomas of the skull base and cervical spine: clinical outcomes associated with a multimodal surgical resection combined with proton-beam radiation in 40 patients. Neurosurg Rev 2012;35:171–82 [discussion 182–3]. [30] Combs SE, Ellerbrock M, Haberer T, et al. Heidelberg Ion Therapy Center (HIT): initial clinical experience in the first 80 patients. Acta Oncol 2010;49:1132–40. [31] Mizoe JE, Hasegawa A, Takagi R, et al. Carbon ion radiotherapy for skull base chordoma. Skull Base 2009;19:219–24.