- Email: [email protected]
MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone
Radiotherapy and Oncology xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Original article
MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone Lorenzo Preda a,b,⇑, Davide Stoppa c, Maria Rosaria Fiore d, Giulia Fontana e, Sofia Camisa c, Roberto Sacchi f, Michele Ghitti f, Gisela Viselner b, Piero Fossati d, Francesca Valvo d, Viviana Vitolo d, Maria Bonora d, Alberto Iannalfi d, Barbara Vischioni d, Alessandro Vai g, Edoardo Mastella g, Guido Baroni h, Roberto Orecchia i,j a
Department of Clinical-Surgical, Diagnostic and Pediatric Sciences, University of Pavia; b Diagnostic Imaging Unit, National Center of Oncological Hadrontherapy (CNAO) Pavia; Diagnostic Radiology Residency School, University of Pavia; d Radiotherapy Unit; e Bio-Engineering Unit, National Center of Oncological Hadrontherapy (CNAO) Pavia; f Applied Statistics Unit, Department of Earth and Environmental Sciences, University of Pavia; g Medical Physics Unit, National Center of Oncological Hadrontherapy (CNAO) Pavia; h Department of Electronics, Information and Bioengineering, Politecnico di Milano; i National Center of Oncological Hadrontherapy (CNAO) Pavia; and j European Institute of Oncology, Milano, Italy c
a r t i c l e
i n f o
Article history: Received 14 September 2017 Received in revised form 20 November 2017 Accepted 30 November 2017 Available online xxxx Keywords: Sacral chordoma Carbon ion radiotherapy RECIST 1.1 Magnetic resonance Diffusion Weighted MRI
a b s t r a c t Background and purpose: To compare RECIST 1.1 with volume modifications in patients with sacral chordoma not suitable for surgery treated with carbon ions radiotherapy (CIRT) alone. To evaluate patients pain before and after CIRT. To detect if baseline Apparent Diffusion Coefficient values (ADC) from Diffusion Weighted sequences could predict response to treatment. Material and methods: Patients included had one cycle of CIRT and underwent MRI before and after treatment. For each MRI, lesion maximum diameter and volume were obtained, and ADC values were analyzed within the whole lesion volume. Patients pain was evaluated with Numerical Rating Scale (NRS), considering the upper tumor level at baseline MRIs. Results: 39 patients were studied (mean follow-up 18 months). Considering RECIST 1.1 there was not a significant reduction in tumor diameters (p = 0.19), instead there was a significant reduction in tumor volume (p < 0.001), with a significant reduction in pain (p = 0.021) if the tumors were above vertebrae S2–S3 at baseline MRIs. The assessment of baseline ADC maps demonstrated higher median values and more negative skewness values in progressive disease (PD) patients versus both partial response (PR) and stable disease (SD). Conclusions: Lesion volume measurement is more accurate than maximum diameter to better stratify the response of sacral chordoma treated with CIRT. Preliminary results suggest that baseline ADC values could be predictive of response to CIRT. Ó 2017 Elsevier B.V. All rights reserved. Radiotherapy and Oncology xxx (2017) xxx–xxx
Chordoma is a rare low-to-intermediate grade bone malignant tumor, which originates from the embryonic notochord’s residues within the axial skeleton [1,2]. This disease represents 1–4% of all primary bone tumors, with an incidence of 0.5–0.8/1,000,000 persons per year [3,4]. Most common localization of the disease is the sacrococcygeal region (50–60%), followed by the clival region (25– 30%), the cervical region (10%) and the thoracolumbar vertebrae (5%) [5]. En bloc surgery with wide resection margins remains the mainstay treatment. Patients who underwent radical resection show both longer local control and disease-free period compared with subtotal resection. However, sacrectomy with wide margins ⇑ Corresponding author at: Diagnostic Imaging Unit, National Center of Oncological Hadrontherapy (CNAO), Strada Privata Campeggi 53, 27100 Pavia, Italy. E-mail address: [email protected] (L. Preda).
is very difficult in large chordoma cases, with a high complication rates especially if performed at S2 level or above, and by the time the first symptoms appear, the tumor may be already too large for surgery [1,6–8]. Hadrontherapy is a form of external beam radiotherapy using beams of charged particles, most commonly protons and carbon ions. It is an effective treatment for sacral chordoma, offering a valid alternative for patients not suitable for surgery [9–11]. In the last decades, hadrontherapy aroused a great interest in clinical application because of dosimetric and radiobiological properties of the particles. The common characteristic of both protons and carbon ions is to release low dose of radiation after penetrating the tissue with a steep dose deposition at the end of their range, called Bragg Peak, which is followed by a steep drop, limiting strongly the dose deposition to surrounding healthy tissues [12]. Furthermore,
https://doi.org/10.1016/j.radonc.2017.11.029 0167-8140/Ó 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
2
MRI of sacral chordoma treated with CIRT
carbon ion beams have a higher relative biological effectiveness compared with protons or X-ray beams [13]. Carbon ions produce multiple damages to the DNA reducing the cell intrinsic repair mechanisms. They are been using mainly for radioresistant tumors arising close to organ at risk to high doses of radiation [14]. Chordoma represents one of the most common tumors treated with carbon ions radiotherapy (CIRT) [12,15]. The purpose of this study was to compare Response Evaluation Criteria in Solid Tumors (RECIST 1.1) [16] with volume modifications for the evaluation of the response in patients with sacral chordoma not suitable for surgery and treated with CIRT alone. Furthermore, we aimed at verifying the correlation of patients’ pain before and after treatment with the dimensional variations of chordomas. Finally, preliminary examinations were performed in order to detect if baseline Apparent Diffusion Coefficient (ADC) values, obtained from Diffusion Weighted Imaging (DWI), could predict response to treatment. Materials and methods Population In this retrospective study we included 39 patients treated with CIRT alone, using active scanning beam delivery system at CNAO (CNAO, Centro Nazionale di Adroterapia Oncologica) of Pavia (Italy) from March 2013 to March 2016. All patients had biopsy proven sacrococcygeal chordoma; they were judged unresectable, and they were consequently treated with a total dose of 70.4 Gy equivalents (GyE) in 16 fraction. Every patient gave an informed consent to the procedure having a personalized treatment plan. Patients with residual sacral chordoma after surgery or with relapse after any other treatments or treated with protons were excluded from the analysis. If the disease relapsed and patients were consequently scheduled for a second cycle of CIRT, only the follow-up exams before the re-irradiation were considered. Imaging protocol and processing Every patient performed a baseline MR exam; during the first year after the end of CIRT, MR examinations were scheduled three-monthly, then biannually. All the MR exams were performed with the same 3 Tesla MR scanner (Magnetom Verio, Siemens Healthineers), using the following sequences: T2-weighted Turbo Spin Echo (TSE) on axial, sagittal and coronal plane, with Fat Saturation (FS); T1-weighted TSE on axial plane; Echo-planar imaging two-dimensional (Ep2d) Diffusion Weighted Images DWI – (b values 50, 400 and 1000 s/mm2) on axial plane; T1-weighted TSE FS on axial, sagittal and coronal plane after intravenous paramagnetic contrast agent administration. For each examination, T2w FS axial images were used both to measure lesion maximum diameter and to get a manual segmentation of the tumor using ITK-SNAP open source software (ver. 3.4.0rcl). The lesion volume was then computed as the integral of the contoured voxels. RECIST 1.1 [16] were used to evaluate chordoma response to CIRT, measuring the maximum lesion diameter on T2w FS axial sequences. Similarly, the same response criteria were applied to volume modifications for treatment response evaluation. In particular, lesions presenting a volume reduction of at least 30% from the
baseline were considered as a partial response; lesions presenting an increase of 20% of it were considered as a progression of disease; lesions between those two limits were considered as stable disease. Furthermore, the segmented lesion was rigidly registered onto the Apparent Diffusion Coefficient (ADC) maps obtained from DWI sequences by means of dedicated Matlab tools (Version: R2010a, The MathWorks, Inc., Natick, MA, USA). In particular, the T2-weighted MR volume was registered onto the ADC map, and the computed transform was applied to warp consequently the segmented lesion. The ADC distribution, within the tumor, was finally characterized by means of the following histogram parameters: median, kurtosis, and skewness values. The Numerical Rating Scale (NRS) was used to measure the painful symptomatology felt by patients, relying on a semiquantitative method based on 11 crescent numeric values, from 0 (total absence of pain) to 10 (the maximum pain imaginable by the patient) [17]. These measures were obtained at baseline evaluation and at the last available follow-up. For this evaluation patients were divided in two subgroups, depending on whether the upper limit of their chordoma was above or below S2–S3 space at baseline MRI. Statistical analysis Linear Mixed Models (LMM) were used to analyze diameter and volume changes during follow-up, in which relative date (number of days from the end of CIRT) was a fixed effect and patient identity was a random effect; this allowed to estimate the variability due only to the patient and not directly to the treatment, improving power and precision of the statistical model. Kaplan–Meier survival analysis was used to estimate rates of local recurrence of disease. Correlation between NRS and the response measured at the end of the follow-up in terms of diameter and volume was analyzed with a covariance analysis (ANCOVA) in which the chordoma upper limit was inserted as a factor. All aforementioned analyses were performed using the software R (ver. 3.3.2, R core Team, 2017) and data reported correspond to means and standard errors. Results 39 patients, 15 females and 24 males with a mean age of 63 years (range 38–84), met the enrollment criteria and were included in the study. They performed baseline MR exams and several follow-up controls after the end of CIRT, from 3 to 37 months afterward (mean time: 18 months), for a total of 195 examinations. RECIST 1.1 vs volumes At baseline examinations, the mean value of maximum axial lesion diameter was 96.9 mm (range: 44–260 mm); the mean lesion volume was 400.6 cm3 (range: 9.1–2418.1 cm3). At the last available follow-up, the mean value of maximum axial lesion diameter was 99.2 mm (range: 46–250 mm); the mean lesion volume was 297.5 cm3 (range: 7.5–2341.4 cm3). LMM analysis showed a significant lesion volume reduction during follow-up (b = 0.035 ± 0.008, Χ2 = 21.089, d.f. = 1, p < 0.001) [Fig. 1a], while there was not a significant variation of maximum diameter (b = 0.0053 ± 0.0041, Χ2 = 1.696, d.f. = 1, p = 0.19) [Fig. 1b]. Patient random effect was highly significant (LR-Χ2 = 74.369, d.f. = 3, p < 0.001) and it explained about 45% of the total variability observed in the volume with the progress of follow-up. Local recurrence was low, as Kaplan-Meier models showed more than 90% of patients free from disease recurrence after six
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
L. Preda et al. / Radiotherapy and Oncology xxx (2017) xxx–xxx
3
changes (287.7 cm3 in baseline MRI and 380.3 cm3 in last followup, +32.2%). 17 patients, 3 PD and 14 SD, did not change their status considering both RECIST 1.1 and volume modifications. None of our patients had a complete response to CIRT. Pain modifications The mean score using the NRS at baseline evaluations was 4.26 (from 0 to 10), while at last available follow-up was 3.21 (from 0 to 8). Particularly, according to the patients stratification obtained from volume changes, in PD patients the mean score varied from 1.25 to 3.25 (3 worsened, 1 stable, 0 improved); in SD patients from 4.29 to 3.21 (2 worsened, 7 stable, 5 improved); in PR patients from 4.81 to 3.19 (4 worsened, 6 stable, 11 improved). Among all 39 patients, 24 had chordoma not reaching S2–S3 sacral intervertebral space at baseline MRIs with their cranial limit, while 15 patients had chordoma with upper cranial limit. ANCOVA highlighted a consensual reduction between NRS values and lesion volume comparing baseline and last follow-up, even if the relationship was at the limits of statistical significance (b = 2. 80 ± 1.76, F = 3.973, d.f. = 1,35, p = 0.054) [Fig. 4a], with a significant reduction in the subgroup of patients with the upper limit of their chordoma above S2–S3 at baseline evaluation (b = 2.44 ± 1.01, F = 5.858, d.f. = 1,35, p = 0.021) [Fig. 4b]. The same analysis repeated considering maximum diameter modifications did not highlight any significant relationship with NRS modifications (F = 0.229, d.f. = 1,34, p = 0.63), but confirmed the effect of the upper lesion limit in baseline evaluation (F = 5.605, d.f. = 1,34, p = 0.024). ADC values
Fig. 1. Treatment response, measured during follow-up, by lesion volume (a) and diameter (b) percentual variation from baseline. The red lines represent the statistical model interpolation (Linear Mixed Model), while the blue lines identify the single patients.
months (according to volume modification: 91 ± 5%; according to RECIST 1.1: 94 ± 4%), and more than 80% at the end of the followup (according to volume modification: 83 ± 9%; according to RECIST 1.1: 88 ± 6%). Considering maximum diameter changes between baseline MRIs and the last available follow-up, according to RECIST 1.1, patients were stratified as follows: 0/39 CR (0%), 0/39 PR (0%), 35/39 SD (89.7%) and 4/39 PD (10.3%). Considering volume changes, applying the same response criteria, patients were classified: 0/39 CR (0%), 21/39 PR (53.8%), 14/39 SD (35.9%) and 4/39 PD (10.3%). 20 patients were considered SD based on maximum diameter changes [Fig. 2], but PR with respect to volume changes [Fig. 3]. 1 patient was considered PD by maximum diameter changes (73 mm in baseline MRI and 96 mm in last follow-up, +31.5%) but PR according to volume changes (66.4 cm3 in baseline MRI and 26.9 cm3 in last follow-up, 59.5%). 1 patient was considered SD by maximum diameter changes (136 mm in baseline MRI and 115 mm in last follow-up, 15.4%) but PD according to volume
ADC maps obtained from DWI sequences were available only in 34 of the 39 baseline MR examinations. According to patients stratification obtained considering volume modifications these 34 patients were classified as 18 PR, 13 SD and 3 PD. The unbalance between PD, PR and SD groups did not allow a proper statistical analysis to assess significant differences, especially between PD and the other two groups. However, higher median ADC values at baseline examinations in PD patients in comparison with both PR and SD patients were highlighted. The median ADC values (mm2/s) of every PD patient were 1653, 1593 and 1737 respectively, while the median of SD patients values was 1166.5 and of PR patients was 1240. Skewness values of every PD patient were respectively 0.04199, 0.57328 and 0.6151, while the median of SD patients values was 0.00133 and of PR patients was 0.41745. Kurtosis values of PD patients were respectively 2.9384, 3.6081 and 3.0613, while the median of SD patients values was 4.4034 and of PR patients was 4.0592. Discussion Surgery remains the gold standard treatment for sacral chordoma, ideally with negative resection margins and generally associated to adjuvant radiotherapy. In patients with localized disease, 5 years survival is about 68%, and the most important negative prognostic factor is positive resection margins, because local failure rates vary approximately between 50% and 100% in subtotal resections, while in en bloc resections they vary between 0% and 53% [18]. A systematic literature review on 40 articles by Pennicooke et al. [19] about toxicity, local control rates and overall survival rates for adjuvant, salvage and primary radiation therapy for spinal and sacral chordoma, concluded that the use of pre- and/or post-operative photon image-guided radiotherapy, proton or carbon ion therapy, should be considered in patients candidate for surgery. In fact such techniques may improve local control rates, while the exclusive use of radiotherapy is still under development
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
4
MRI of sacral chordoma treated with CIRT
Fig. 2. Male, 67 years old patient. TSE T2 FS images on axial and sagittal planes from MR exams, acquired respectively at baseline, at one year follow-up and at two years follow-up (from left to right): according to RECIST 1.1, 124 mm vs 88 mm, 29% (classified as SD).
Fig. 3. Same patient of Fig. 2. Tridimensional lesion volume renderings on axial, coronal and sagittal planes, realized from manual segmentations obtained from the same MR exams mentioned: 868.5 cm3 vs 136.2 cm3, 82.8% (classified as PR).
since it shows good results but it is necessary a clear delimitation of toxicity profiles and long-term local control. Currently there is only a very limited number of centers around the world which offers the possibility to treat chordoma using CIRT. It offers advantages both as adjuvant treatment to surgery and as a primary treatment [11,20]. Uhl et al. [20] reported the first 56 patients treated for sacral chordoma with CIRT alone or after surgery for primitive tumor and local recurrence, showing local control rates at 2 and 3 years respectively of 76% and 53%, with a global survival rate of 100% with no higher toxicity within the mean follow-up time of 25 months. Imai et. al. [11] reported long term results on a larger group of 188 patients with inoperable sacral chordoma, all treated with CIRT only. After a median follow-up time of 62 months, reported 5 years local control, overall
survival and disease free survival rates were respectively 77.2%, 81.1% and 50.3%. The evaluation of the therapy response is of utmost importance, in order to establish its effectiveness in terms of local control with low toxicity. All patients receiving diagnosis of sacral chordoma when the tumor is not suitable for surgery need a valid therapeutic alternative. Furthermore, it is helpful to try to standardize the response assessment method, in order to avoid excessive variability among different operators and even risk to misunderstand the results. An approach to this problem has been proposed with RECIST (updated to 1.1 version in 2009) [16], an universally accepted method of linear measurement which account among its main advantages the ease of execution and high intra-and inter-observer reproducibility. Measuring the entire volume of
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
L. Preda et al. / Radiotherapy and Oncology xxx (2017) xxx–xxx
Fig. 4. NRS variation at the end of follow-up, according to lesion volume modifications (a) and depending on whether the upper limit of chordoma was below or above S2–S3 space at baseline MR evaluations (b).
the lesion is a less quick but more accurate method, since it provides a more reliable surrogate of the entire lesion extension when compared to the mono-dimensional measure derived by the diameter. This has recently been reported by Kabolizadeh et al. in 2017 [18], who described a retrospective analysis conducted on 40 patients with unresected spinal chordoma treated by photon/proton radiotherapy, of which only 19 had a complete set of images, with a mean follow-up of 50.3 months, reporting that volumetric analysis, compared to RECIST 1.1, was more reliable and reproducible, helping to measure even minimal changes in the size of the tumors. In our study, we have achieved similar results: using RECIST 1.1, virtually all patients would be classified as stable disease and none in response, while more than 50% of patients had significant reduction in tumor volume. Consequently the measurement of the maximum diameter of sacral chordoma does not seem sufficient to fully describe lesion modifications after CIRT, especially because
5
of progressive and irregular shrinking, which is not properly reflected on their maximum axial diameters in most cases. Response evaluation by RECIST 1.1 is thus less sensitive when compared to the complete lesion volumetric assessment. Although it involves more time consuming procedure, it was reported to be a more appropriate approach toward sacral chordoma changes definition after CIRT thus allowing reliable patients stratification in different response categories at earlier follow-up. Another important parameter in evaluating treatment response is the improvement of patient’s quality of life. In our study we chose to use a proven and effective method of obtaining a semiquantitative measure of pain, complained by almost all of our patients, such as the NRS. A correlation was found between tumor dimensional variations and variations in NRS values, assessed at baseline and at the latest available follow-up. There was a correlation at limits of statistical significance (p = 0.054) between NRS reduction and tumor volume reduction, while there was no significant correlation with diameter variations (p = 0.63). However, this correlation was strongly influenced by the upper limit of the lesion, in fact it reached statistical significance only in the subgroup of patients whose sacral chordoma had the upper limit above the S2–S3 intervertebral space at the baseline evaluation, both considering volumetric (p = 0.021) and diameter (p = 0.024) variations. This finding agrees with the literature data [6,11], indicating that a possible sacrectomy above S2 is more related to functional limitations and neurological damage due to the more extensive involvement of the nerve roots of the sacral plexus. Again, the evaluation of sacral chordoma by measuring their entire volume has allowed to obtain a more accurate stratification of the response to CIRT. In addition to conventional MR sequences, useful for the morphological investigation of the lesion, there is an increasing interest in DWI, which can provide indirect information about the microstructural characteristics of the tissue. This is a wellestablished MR technique, originally designed for early diagnosis of brain stroke, which quantifies the Brownian motion of water molecules in tissues by calculating the ADC. It is increasingly used in many oncological applications to obtain indirect information about tumor tissues; particularly, it was reported to be an effective tool in distinguishing benign lesions from malignant lesions and in evaluating the response to (chemo-) radiotherapy [21]. To the best of our knowledge, there are currently no studies regarding the possibility of predicting the response to the treatment of sacral chordoma using the information derived from DW-MR sequences. In our study, non-responder patients had baseline ADC values higher than patients with stable disease and those in partial response, with also more negative skewness values compared to stable patients and patients in partial response. These preliminary findings may be indicative of a different internal microstructure of the lesion, as already described for other tumor models where authors found a positive correlation between ADC values and tumor stromal component as a negative prognostic factor [22]. Although the limited number of non-responder patients did not allow a proper statistical evaluation in the present study, the gathered results suggest potential biomarkers of treatment response to be found among baseline ADC values. Thus, further investigations including larger population would be greatly useful to assess such DWI-MR potentiality for sacrococcygeal chordoma, in particular to identify potential non-responsive patients, anticipating their dimensional progression over time. The main limitations of this study are the low number of patients, partially justified by the rarity of the pathology, the elite treatment performed, and the heterogeneous follow-up period available. It would be desirable a longer observation period, possibly not less than 24–36 months [11,18,20]. In addition, the percentage of patients who had recurrence of disease is lower in
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
6
MRI of sacral chordoma treated with CIRT
comparison with patients with stable disease or in partial response; consequently, observations regarding the potential predictive role of ADC values should be considered completely preliminary. In conclusion our findings are consistent with what has been observed in other studies, thus supporting the thesis that the evaluation of the response of sacral chordoma to radiotherapy should be performed by measuring their entire volumes, instead of relying on conventional RECIST 1.1 based on mono-dimensional lesion measurement. Lesion volume reported more accurate and earlier stratification of patients capability, thus resulting in a more precise description of the response to treatment. This evaluation also seems to correlate better with clinical outcomes, based on the evaluation of pain through NRS. Preliminary results of our study also suggest that ADC values in baseline MRI examinations may predict the response to CIRT, in particular to identify potential nonresponder patients. These observations have to be confirmed by future studies including a wider population. Conflict of interest None. References [1] Kayani B, Hanna SA, Sewell MD, Saifuddin A, Molloy S, Briggs TWR. A review of the surgical management of sacral chordoma. Eur J SurgOncol EJSO 2014;40:1412–20. [2] Jemal A, Siegel R, Xu J, Ward E. Cancer statistics. CA Cancer J Clin 2010;60:277–300. https://doi.org/10.3322/caac.20073. [3] Eriksson B, Gunterberg B, Kindblom LG. Chordoma: a clinicopathologic and prognostic study of a Swedish national series. Acta OrthopScand 1981;52:49–58. [4] McMaster ML, Goldstein AM, Bromley CM, et al. Chordoma: incidence and survival patterns in the United States, 1973–1995. Cancer Causes Control 2001;12:1–11. [5] Chugh R, Tawbi H, Lucas DR, Biermann JS, Schuetze SM, Baker LH. Chordoma: the nonsarcoma primary bone tumor. Oncologist 2007;12:1344–50. https:// doi.org/10.1634/theoncologist.12-11-1344.
[6] Cheng EY, Ozerdemoglu RA, Transfeldt EE, et al. Lumbosacral chordoma: Prognostic factors and treatment. Spine 1999;24:1639–45. [7] Boriani S, Chevalley F, Weinstein JN, et al. Chordoma of the spine above the sacrum: Treatment and outcome in 21 cases. Spine 1996;21:1569–77. [8] Catton C, O’Sullivan B, Bell R, et al. Chordoma: long-term follow-up after radical photon irradiation. Radiother Oncol 1996;41:67–72. [9] Imai R, Kamada T, Tsuji H, et al. Effect of carbon ion radiotherapy for sacral chordoma: results of Phase I-II and Phase II clinical trials. Int J Radiat Oncol Biol Phys 2010;77:1470–6. [10] Nishida Y, Kamada T, Imai R, et al. Clinical outcome of sacral chordoma with carbon ion radiotherapy compared with surgery. Int J Radiat Oncol Biol Phys 2011;79:110–6. [11] Imai R, Kamada T, Araki N, et al. Carbon ion radiation therapy for unresectable sacral chordoma: an analysis of 188 cases. Int J Radiat Oncol Biol Phys 2016 May 1;95:322–7. [12] Imai R, Kamada T, Tsuji H, et al. Carbon ion radiotherapy for unresectable sacral chordomas. Clin Cancer Res 2004;10:5741–6. [13] Ohno Tatsuya. Particle radiotherapy with carbon ion beams. EPMA J 2013;4:9. [14] Kraft G. The radiobiological and physical basis for radiotherapy with protons and heavier ions. Strahlenther Onkol 1990;166:10. [15] Kamada T, Tsujii H, Tsuji H, et al. Efficacy and safety of CIRT in bone and soft tissue sarcomas. J Clin Oncol 2002;20:4466–71. [16] Eisenhauera EA, Therasseb P, Bogaertsc J, et al. New response evaluation criteria in solid tumors: Revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228–47. [17] Beebe A, McCaffery M, et al. Pain: clinical manual for nursing practice. Louis, MO: Mosby St; 1989. [18] Kabolizadeh Peyman, Che YL, Liebsch N, et al. Updated outcome and analysis of tumor response in mobile spine and sacral chordoma treated with definitive high-dose photon/proton radiation therapy. Int J Radiat Oncol Biol Phys 2017;97:254–62. [19] Pennicooke B, Laufer I, Sahgal A, Varga PP, Gokaslan ZL, Bilsky MH, et al. Safety and local control of radiation therapy for chordoma of the spine and sacrum: a systematic review. Spine (PhilaPa 1976) 2016;41:S186–92. [20] Uhl M, Welzel T, Jensen A, Ellerbrock M, Haberer T, Jäkel O, et al. Carbon ion beam treatment in patients with primary and recurrent sacrococcygeal chordoma. Strahlenther Onkol 2015;191:597–603. 10.1007. [21] Taouli B, Beer AJ, Chenevert T, Collins D, Lehman C, Matos C, et al. Koh DM Diffusion-weighted imaging outside the brain: Consensus statement from an ISMRM-sponsored workshop. J Magn Reson Imaging 2016;44:521–40. https:// doi.org/10.1002/jmri.25196. Epub 2016 Feb 19. [22] Driessen JP, Caldas-Magalhaes J, Janssen LM, et al. Diffusion-weighted MR imaging in laryngeal and hypopharyngeal carcinoma: association between apparent diffusion coefficient and histologic findings. Radiology 2014;272:456–63.
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
Contents lists available at ScienceDirect
Radiotherapy and Oncology journal homepage: www.thegreenjournal.com
Original article
MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone Lorenzo Preda a,b,⇑, Davide Stoppa c, Maria Rosaria Fiore d, Giulia Fontana e, Sofia Camisa c, Roberto Sacchi f, Michele Ghitti f, Gisela Viselner b, Piero Fossati d, Francesca Valvo d, Viviana Vitolo d, Maria Bonora d, Alberto Iannalfi d, Barbara Vischioni d, Alessandro Vai g, Edoardo Mastella g, Guido Baroni h, Roberto Orecchia i,j a
Department of Clinical-Surgical, Diagnostic and Pediatric Sciences, University of Pavia; b Diagnostic Imaging Unit, National Center of Oncological Hadrontherapy (CNAO) Pavia; Diagnostic Radiology Residency School, University of Pavia; d Radiotherapy Unit; e Bio-Engineering Unit, National Center of Oncological Hadrontherapy (CNAO) Pavia; f Applied Statistics Unit, Department of Earth and Environmental Sciences, University of Pavia; g Medical Physics Unit, National Center of Oncological Hadrontherapy (CNAO) Pavia; h Department of Electronics, Information and Bioengineering, Politecnico di Milano; i National Center of Oncological Hadrontherapy (CNAO) Pavia; and j European Institute of Oncology, Milano, Italy c
a r t i c l e
i n f o
Article history: Received 14 September 2017 Received in revised form 20 November 2017 Accepted 30 November 2017 Available online xxxx Keywords: Sacral chordoma Carbon ion radiotherapy RECIST 1.1 Magnetic resonance Diffusion Weighted MRI
a b s t r a c t Background and purpose: To compare RECIST 1.1 with volume modifications in patients with sacral chordoma not suitable for surgery treated with carbon ions radiotherapy (CIRT) alone. To evaluate patients pain before and after CIRT. To detect if baseline Apparent Diffusion Coefficient values (ADC) from Diffusion Weighted sequences could predict response to treatment. Material and methods: Patients included had one cycle of CIRT and underwent MRI before and after treatment. For each MRI, lesion maximum diameter and volume were obtained, and ADC values were analyzed within the whole lesion volume. Patients pain was evaluated with Numerical Rating Scale (NRS), considering the upper tumor level at baseline MRIs. Results: 39 patients were studied (mean follow-up 18 months). Considering RECIST 1.1 there was not a significant reduction in tumor diameters (p = 0.19), instead there was a significant reduction in tumor volume (p < 0.001), with a significant reduction in pain (p = 0.021) if the tumors were above vertebrae S2–S3 at baseline MRIs. The assessment of baseline ADC maps demonstrated higher median values and more negative skewness values in progressive disease (PD) patients versus both partial response (PR) and stable disease (SD). Conclusions: Lesion volume measurement is more accurate than maximum diameter to better stratify the response of sacral chordoma treated with CIRT. Preliminary results suggest that baseline ADC values could be predictive of response to CIRT. Ó 2017 Elsevier B.V. All rights reserved. Radiotherapy and Oncology xxx (2017) xxx–xxx
Chordoma is a rare low-to-intermediate grade bone malignant tumor, which originates from the embryonic notochord’s residues within the axial skeleton [1,2]. This disease represents 1–4% of all primary bone tumors, with an incidence of 0.5–0.8/1,000,000 persons per year [3,4]. Most common localization of the disease is the sacrococcygeal region (50–60%), followed by the clival region (25– 30%), the cervical region (10%) and the thoracolumbar vertebrae (5%) [5]. En bloc surgery with wide resection margins remains the mainstay treatment. Patients who underwent radical resection show both longer local control and disease-free period compared with subtotal resection. However, sacrectomy with wide margins ⇑ Corresponding author at: Diagnostic Imaging Unit, National Center of Oncological Hadrontherapy (CNAO), Strada Privata Campeggi 53, 27100 Pavia, Italy. E-mail address: [email protected] (L. Preda).
is very difficult in large chordoma cases, with a high complication rates especially if performed at S2 level or above, and by the time the first symptoms appear, the tumor may be already too large for surgery [1,6–8]. Hadrontherapy is a form of external beam radiotherapy using beams of charged particles, most commonly protons and carbon ions. It is an effective treatment for sacral chordoma, offering a valid alternative for patients not suitable for surgery [9–11]. In the last decades, hadrontherapy aroused a great interest in clinical application because of dosimetric and radiobiological properties of the particles. The common characteristic of both protons and carbon ions is to release low dose of radiation after penetrating the tissue with a steep dose deposition at the end of their range, called Bragg Peak, which is followed by a steep drop, limiting strongly the dose deposition to surrounding healthy tissues [12]. Furthermore,
https://doi.org/10.1016/j.radonc.2017.11.029 0167-8140/Ó 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
2
MRI of sacral chordoma treated with CIRT
carbon ion beams have a higher relative biological effectiveness compared with protons or X-ray beams [13]. Carbon ions produce multiple damages to the DNA reducing the cell intrinsic repair mechanisms. They are been using mainly for radioresistant tumors arising close to organ at risk to high doses of radiation [14]. Chordoma represents one of the most common tumors treated with carbon ions radiotherapy (CIRT) [12,15]. The purpose of this study was to compare Response Evaluation Criteria in Solid Tumors (RECIST 1.1) [16] with volume modifications for the evaluation of the response in patients with sacral chordoma not suitable for surgery and treated with CIRT alone. Furthermore, we aimed at verifying the correlation of patients’ pain before and after treatment with the dimensional variations of chordomas. Finally, preliminary examinations were performed in order to detect if baseline Apparent Diffusion Coefficient (ADC) values, obtained from Diffusion Weighted Imaging (DWI), could predict response to treatment. Materials and methods Population In this retrospective study we included 39 patients treated with CIRT alone, using active scanning beam delivery system at CNAO (CNAO, Centro Nazionale di Adroterapia Oncologica) of Pavia (Italy) from March 2013 to March 2016. All patients had biopsy proven sacrococcygeal chordoma; they were judged unresectable, and they were consequently treated with a total dose of 70.4 Gy equivalents (GyE) in 16 fraction. Every patient gave an informed consent to the procedure having a personalized treatment plan. Patients with residual sacral chordoma after surgery or with relapse after any other treatments or treated with protons were excluded from the analysis. If the disease relapsed and patients were consequently scheduled for a second cycle of CIRT, only the follow-up exams before the re-irradiation were considered. Imaging protocol and processing Every patient performed a baseline MR exam; during the first year after the end of CIRT, MR examinations were scheduled three-monthly, then biannually. All the MR exams were performed with the same 3 Tesla MR scanner (Magnetom Verio, Siemens Healthineers), using the following sequences: T2-weighted Turbo Spin Echo (TSE) on axial, sagittal and coronal plane, with Fat Saturation (FS); T1-weighted TSE on axial plane; Echo-planar imaging two-dimensional (Ep2d) Diffusion Weighted Images DWI – (b values 50, 400 and 1000 s/mm2) on axial plane; T1-weighted TSE FS on axial, sagittal and coronal plane after intravenous paramagnetic contrast agent administration. For each examination, T2w FS axial images were used both to measure lesion maximum diameter and to get a manual segmentation of the tumor using ITK-SNAP open source software (ver. 3.4.0rcl). The lesion volume was then computed as the integral of the contoured voxels. RECIST 1.1 [16] were used to evaluate chordoma response to CIRT, measuring the maximum lesion diameter on T2w FS axial sequences. Similarly, the same response criteria were applied to volume modifications for treatment response evaluation. In particular, lesions presenting a volume reduction of at least 30% from the
baseline were considered as a partial response; lesions presenting an increase of 20% of it were considered as a progression of disease; lesions between those two limits were considered as stable disease. Furthermore, the segmented lesion was rigidly registered onto the Apparent Diffusion Coefficient (ADC) maps obtained from DWI sequences by means of dedicated Matlab tools (Version: R2010a, The MathWorks, Inc., Natick, MA, USA). In particular, the T2-weighted MR volume was registered onto the ADC map, and the computed transform was applied to warp consequently the segmented lesion. The ADC distribution, within the tumor, was finally characterized by means of the following histogram parameters: median, kurtosis, and skewness values. The Numerical Rating Scale (NRS) was used to measure the painful symptomatology felt by patients, relying on a semiquantitative method based on 11 crescent numeric values, from 0 (total absence of pain) to 10 (the maximum pain imaginable by the patient) [17]. These measures were obtained at baseline evaluation and at the last available follow-up. For this evaluation patients were divided in two subgroups, depending on whether the upper limit of their chordoma was above or below S2–S3 space at baseline MRI. Statistical analysis Linear Mixed Models (LMM) were used to analyze diameter and volume changes during follow-up, in which relative date (number of days from the end of CIRT) was a fixed effect and patient identity was a random effect; this allowed to estimate the variability due only to the patient and not directly to the treatment, improving power and precision of the statistical model. Kaplan–Meier survival analysis was used to estimate rates of local recurrence of disease. Correlation between NRS and the response measured at the end of the follow-up in terms of diameter and volume was analyzed with a covariance analysis (ANCOVA) in which the chordoma upper limit was inserted as a factor. All aforementioned analyses were performed using the software R (ver. 3.3.2, R core Team, 2017) and data reported correspond to means and standard errors. Results 39 patients, 15 females and 24 males with a mean age of 63 years (range 38–84), met the enrollment criteria and were included in the study. They performed baseline MR exams and several follow-up controls after the end of CIRT, from 3 to 37 months afterward (mean time: 18 months), for a total of 195 examinations. RECIST 1.1 vs volumes At baseline examinations, the mean value of maximum axial lesion diameter was 96.9 mm (range: 44–260 mm); the mean lesion volume was 400.6 cm3 (range: 9.1–2418.1 cm3). At the last available follow-up, the mean value of maximum axial lesion diameter was 99.2 mm (range: 46–250 mm); the mean lesion volume was 297.5 cm3 (range: 7.5–2341.4 cm3). LMM analysis showed a significant lesion volume reduction during follow-up (b = 0.035 ± 0.008, Χ2 = 21.089, d.f. = 1, p < 0.001) [Fig. 1a], while there was not a significant variation of maximum diameter (b = 0.0053 ± 0.0041, Χ2 = 1.696, d.f. = 1, p = 0.19) [Fig. 1b]. Patient random effect was highly significant (LR-Χ2 = 74.369, d.f. = 3, p < 0.001) and it explained about 45% of the total variability observed in the volume with the progress of follow-up. Local recurrence was low, as Kaplan-Meier models showed more than 90% of patients free from disease recurrence after six
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
L. Preda et al. / Radiotherapy and Oncology xxx (2017) xxx–xxx
3
changes (287.7 cm3 in baseline MRI and 380.3 cm3 in last followup, +32.2%). 17 patients, 3 PD and 14 SD, did not change their status considering both RECIST 1.1 and volume modifications. None of our patients had a complete response to CIRT. Pain modifications The mean score using the NRS at baseline evaluations was 4.26 (from 0 to 10), while at last available follow-up was 3.21 (from 0 to 8). Particularly, according to the patients stratification obtained from volume changes, in PD patients the mean score varied from 1.25 to 3.25 (3 worsened, 1 stable, 0 improved); in SD patients from 4.29 to 3.21 (2 worsened, 7 stable, 5 improved); in PR patients from 4.81 to 3.19 (4 worsened, 6 stable, 11 improved). Among all 39 patients, 24 had chordoma not reaching S2–S3 sacral intervertebral space at baseline MRIs with their cranial limit, while 15 patients had chordoma with upper cranial limit. ANCOVA highlighted a consensual reduction between NRS values and lesion volume comparing baseline and last follow-up, even if the relationship was at the limits of statistical significance (b = 2. 80 ± 1.76, F = 3.973, d.f. = 1,35, p = 0.054) [Fig. 4a], with a significant reduction in the subgroup of patients with the upper limit of their chordoma above S2–S3 at baseline evaluation (b = 2.44 ± 1.01, F = 5.858, d.f. = 1,35, p = 0.021) [Fig. 4b]. The same analysis repeated considering maximum diameter modifications did not highlight any significant relationship with NRS modifications (F = 0.229, d.f. = 1,34, p = 0.63), but confirmed the effect of the upper lesion limit in baseline evaluation (F = 5.605, d.f. = 1,34, p = 0.024). ADC values
Fig. 1. Treatment response, measured during follow-up, by lesion volume (a) and diameter (b) percentual variation from baseline. The red lines represent the statistical model interpolation (Linear Mixed Model), while the blue lines identify the single patients.
months (according to volume modification: 91 ± 5%; according to RECIST 1.1: 94 ± 4%), and more than 80% at the end of the followup (according to volume modification: 83 ± 9%; according to RECIST 1.1: 88 ± 6%). Considering maximum diameter changes between baseline MRIs and the last available follow-up, according to RECIST 1.1, patients were stratified as follows: 0/39 CR (0%), 0/39 PR (0%), 35/39 SD (89.7%) and 4/39 PD (10.3%). Considering volume changes, applying the same response criteria, patients were classified: 0/39 CR (0%), 21/39 PR (53.8%), 14/39 SD (35.9%) and 4/39 PD (10.3%). 20 patients were considered SD based on maximum diameter changes [Fig. 2], but PR with respect to volume changes [Fig. 3]. 1 patient was considered PD by maximum diameter changes (73 mm in baseline MRI and 96 mm in last follow-up, +31.5%) but PR according to volume changes (66.4 cm3 in baseline MRI and 26.9 cm3 in last follow-up, 59.5%). 1 patient was considered SD by maximum diameter changes (136 mm in baseline MRI and 115 mm in last follow-up, 15.4%) but PD according to volume
ADC maps obtained from DWI sequences were available only in 34 of the 39 baseline MR examinations. According to patients stratification obtained considering volume modifications these 34 patients were classified as 18 PR, 13 SD and 3 PD. The unbalance between PD, PR and SD groups did not allow a proper statistical analysis to assess significant differences, especially between PD and the other two groups. However, higher median ADC values at baseline examinations in PD patients in comparison with both PR and SD patients were highlighted. The median ADC values (mm2/s) of every PD patient were 1653, 1593 and 1737 respectively, while the median of SD patients values was 1166.5 and of PR patients was 1240. Skewness values of every PD patient were respectively 0.04199, 0.57328 and 0.6151, while the median of SD patients values was 0.00133 and of PR patients was 0.41745. Kurtosis values of PD patients were respectively 2.9384, 3.6081 and 3.0613, while the median of SD patients values was 4.4034 and of PR patients was 4.0592. Discussion Surgery remains the gold standard treatment for sacral chordoma, ideally with negative resection margins and generally associated to adjuvant radiotherapy. In patients with localized disease, 5 years survival is about 68%, and the most important negative prognostic factor is positive resection margins, because local failure rates vary approximately between 50% and 100% in subtotal resections, while in en bloc resections they vary between 0% and 53% [18]. A systematic literature review on 40 articles by Pennicooke et al. [19] about toxicity, local control rates and overall survival rates for adjuvant, salvage and primary radiation therapy for spinal and sacral chordoma, concluded that the use of pre- and/or post-operative photon image-guided radiotherapy, proton or carbon ion therapy, should be considered in patients candidate for surgery. In fact such techniques may improve local control rates, while the exclusive use of radiotherapy is still under development
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
4
MRI of sacral chordoma treated with CIRT
Fig. 2. Male, 67 years old patient. TSE T2 FS images on axial and sagittal planes from MR exams, acquired respectively at baseline, at one year follow-up and at two years follow-up (from left to right): according to RECIST 1.1, 124 mm vs 88 mm, 29% (classified as SD).
Fig. 3. Same patient of Fig. 2. Tridimensional lesion volume renderings on axial, coronal and sagittal planes, realized from manual segmentations obtained from the same MR exams mentioned: 868.5 cm3 vs 136.2 cm3, 82.8% (classified as PR).
since it shows good results but it is necessary a clear delimitation of toxicity profiles and long-term local control. Currently there is only a very limited number of centers around the world which offers the possibility to treat chordoma using CIRT. It offers advantages both as adjuvant treatment to surgery and as a primary treatment [11,20]. Uhl et al. [20] reported the first 56 patients treated for sacral chordoma with CIRT alone or after surgery for primitive tumor and local recurrence, showing local control rates at 2 and 3 years respectively of 76% and 53%, with a global survival rate of 100% with no higher toxicity within the mean follow-up time of 25 months. Imai et. al. [11] reported long term results on a larger group of 188 patients with inoperable sacral chordoma, all treated with CIRT only. After a median follow-up time of 62 months, reported 5 years local control, overall
survival and disease free survival rates were respectively 77.2%, 81.1% and 50.3%. The evaluation of the therapy response is of utmost importance, in order to establish its effectiveness in terms of local control with low toxicity. All patients receiving diagnosis of sacral chordoma when the tumor is not suitable for surgery need a valid therapeutic alternative. Furthermore, it is helpful to try to standardize the response assessment method, in order to avoid excessive variability among different operators and even risk to misunderstand the results. An approach to this problem has been proposed with RECIST (updated to 1.1 version in 2009) [16], an universally accepted method of linear measurement which account among its main advantages the ease of execution and high intra-and inter-observer reproducibility. Measuring the entire volume of
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
L. Preda et al. / Radiotherapy and Oncology xxx (2017) xxx–xxx
Fig. 4. NRS variation at the end of follow-up, according to lesion volume modifications (a) and depending on whether the upper limit of chordoma was below or above S2–S3 space at baseline MR evaluations (b).
the lesion is a less quick but more accurate method, since it provides a more reliable surrogate of the entire lesion extension when compared to the mono-dimensional measure derived by the diameter. This has recently been reported by Kabolizadeh et al. in 2017 [18], who described a retrospective analysis conducted on 40 patients with unresected spinal chordoma treated by photon/proton radiotherapy, of which only 19 had a complete set of images, with a mean follow-up of 50.3 months, reporting that volumetric analysis, compared to RECIST 1.1, was more reliable and reproducible, helping to measure even minimal changes in the size of the tumors. In our study, we have achieved similar results: using RECIST 1.1, virtually all patients would be classified as stable disease and none in response, while more than 50% of patients had significant reduction in tumor volume. Consequently the measurement of the maximum diameter of sacral chordoma does not seem sufficient to fully describe lesion modifications after CIRT, especially because
5
of progressive and irregular shrinking, which is not properly reflected on their maximum axial diameters in most cases. Response evaluation by RECIST 1.1 is thus less sensitive when compared to the complete lesion volumetric assessment. Although it involves more time consuming procedure, it was reported to be a more appropriate approach toward sacral chordoma changes definition after CIRT thus allowing reliable patients stratification in different response categories at earlier follow-up. Another important parameter in evaluating treatment response is the improvement of patient’s quality of life. In our study we chose to use a proven and effective method of obtaining a semiquantitative measure of pain, complained by almost all of our patients, such as the NRS. A correlation was found between tumor dimensional variations and variations in NRS values, assessed at baseline and at the latest available follow-up. There was a correlation at limits of statistical significance (p = 0.054) between NRS reduction and tumor volume reduction, while there was no significant correlation with diameter variations (p = 0.63). However, this correlation was strongly influenced by the upper limit of the lesion, in fact it reached statistical significance only in the subgroup of patients whose sacral chordoma had the upper limit above the S2–S3 intervertebral space at the baseline evaluation, both considering volumetric (p = 0.021) and diameter (p = 0.024) variations. This finding agrees with the literature data [6,11], indicating that a possible sacrectomy above S2 is more related to functional limitations and neurological damage due to the more extensive involvement of the nerve roots of the sacral plexus. Again, the evaluation of sacral chordoma by measuring their entire volume has allowed to obtain a more accurate stratification of the response to CIRT. In addition to conventional MR sequences, useful for the morphological investigation of the lesion, there is an increasing interest in DWI, which can provide indirect information about the microstructural characteristics of the tissue. This is a wellestablished MR technique, originally designed for early diagnosis of brain stroke, which quantifies the Brownian motion of water molecules in tissues by calculating the ADC. It is increasingly used in many oncological applications to obtain indirect information about tumor tissues; particularly, it was reported to be an effective tool in distinguishing benign lesions from malignant lesions and in evaluating the response to (chemo-) radiotherapy [21]. To the best of our knowledge, there are currently no studies regarding the possibility of predicting the response to the treatment of sacral chordoma using the information derived from DW-MR sequences. In our study, non-responder patients had baseline ADC values higher than patients with stable disease and those in partial response, with also more negative skewness values compared to stable patients and patients in partial response. These preliminary findings may be indicative of a different internal microstructure of the lesion, as already described for other tumor models where authors found a positive correlation between ADC values and tumor stromal component as a negative prognostic factor [22]. Although the limited number of non-responder patients did not allow a proper statistical evaluation in the present study, the gathered results suggest potential biomarkers of treatment response to be found among baseline ADC values. Thus, further investigations including larger population would be greatly useful to assess such DWI-MR potentiality for sacrococcygeal chordoma, in particular to identify potential non-responsive patients, anticipating their dimensional progression over time. The main limitations of this study are the low number of patients, partially justified by the rarity of the pathology, the elite treatment performed, and the heterogeneous follow-up period available. It would be desirable a longer observation period, possibly not less than 24–36 months [11,18,20]. In addition, the percentage of patients who had recurrence of disease is lower in
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029
6
MRI of sacral chordoma treated with CIRT
comparison with patients with stable disease or in partial response; consequently, observations regarding the potential predictive role of ADC values should be considered completely preliminary. In conclusion our findings are consistent with what has been observed in other studies, thus supporting the thesis that the evaluation of the response of sacral chordoma to radiotherapy should be performed by measuring their entire volumes, instead of relying on conventional RECIST 1.1 based on mono-dimensional lesion measurement. Lesion volume reported more accurate and earlier stratification of patients capability, thus resulting in a more precise description of the response to treatment. This evaluation also seems to correlate better with clinical outcomes, based on the evaluation of pain through NRS. Preliminary results of our study also suggest that ADC values in baseline MRI examinations may predict the response to CIRT, in particular to identify potential nonresponder patients. These observations have to be confirmed by future studies including a wider population. Conflict of interest None. References [1] Kayani B, Hanna SA, Sewell MD, Saifuddin A, Molloy S, Briggs TWR. A review of the surgical management of sacral chordoma. Eur J SurgOncol EJSO 2014;40:1412–20. [2] Jemal A, Siegel R, Xu J, Ward E. Cancer statistics. CA Cancer J Clin 2010;60:277–300. https://doi.org/10.3322/caac.20073. [3] Eriksson B, Gunterberg B, Kindblom LG. Chordoma: a clinicopathologic and prognostic study of a Swedish national series. Acta OrthopScand 1981;52:49–58. [4] McMaster ML, Goldstein AM, Bromley CM, et al. Chordoma: incidence and survival patterns in the United States, 1973–1995. Cancer Causes Control 2001;12:1–11. [5] Chugh R, Tawbi H, Lucas DR, Biermann JS, Schuetze SM, Baker LH. Chordoma: the nonsarcoma primary bone tumor. Oncologist 2007;12:1344–50. https:// doi.org/10.1634/theoncologist.12-11-1344.
[6] Cheng EY, Ozerdemoglu RA, Transfeldt EE, et al. Lumbosacral chordoma: Prognostic factors and treatment. Spine 1999;24:1639–45. [7] Boriani S, Chevalley F, Weinstein JN, et al. Chordoma of the spine above the sacrum: Treatment and outcome in 21 cases. Spine 1996;21:1569–77. [8] Catton C, O’Sullivan B, Bell R, et al. Chordoma: long-term follow-up after radical photon irradiation. Radiother Oncol 1996;41:67–72. [9] Imai R, Kamada T, Tsuji H, et al. Effect of carbon ion radiotherapy for sacral chordoma: results of Phase I-II and Phase II clinical trials. Int J Radiat Oncol Biol Phys 2010;77:1470–6. [10] Nishida Y, Kamada T, Imai R, et al. Clinical outcome of sacral chordoma with carbon ion radiotherapy compared with surgery. Int J Radiat Oncol Biol Phys 2011;79:110–6. [11] Imai R, Kamada T, Araki N, et al. Carbon ion radiation therapy for unresectable sacral chordoma: an analysis of 188 cases. Int J Radiat Oncol Biol Phys 2016 May 1;95:322–7. [12] Imai R, Kamada T, Tsuji H, et al. Carbon ion radiotherapy for unresectable sacral chordomas. Clin Cancer Res 2004;10:5741–6. [13] Ohno Tatsuya. Particle radiotherapy with carbon ion beams. EPMA J 2013;4:9. [14] Kraft G. The radiobiological and physical basis for radiotherapy with protons and heavier ions. Strahlenther Onkol 1990;166:10. [15] Kamada T, Tsujii H, Tsuji H, et al. Efficacy and safety of CIRT in bone and soft tissue sarcomas. J Clin Oncol 2002;20:4466–71. [16] Eisenhauera EA, Therasseb P, Bogaertsc J, et al. New response evaluation criteria in solid tumors: Revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228–47. [17] Beebe A, McCaffery M, et al. Pain: clinical manual for nursing practice. Louis, MO: Mosby St; 1989. [18] Kabolizadeh Peyman, Che YL, Liebsch N, et al. Updated outcome and analysis of tumor response in mobile spine and sacral chordoma treated with definitive high-dose photon/proton radiation therapy. Int J Radiat Oncol Biol Phys 2017;97:254–62. [19] Pennicooke B, Laufer I, Sahgal A, Varga PP, Gokaslan ZL, Bilsky MH, et al. Safety and local control of radiation therapy for chordoma of the spine and sacrum: a systematic review. Spine (PhilaPa 1976) 2016;41:S186–92. [20] Uhl M, Welzel T, Jensen A, Ellerbrock M, Haberer T, Jäkel O, et al. Carbon ion beam treatment in patients with primary and recurrent sacrococcygeal chordoma. Strahlenther Onkol 2015;191:597–603. 10.1007. [21] Taouli B, Beer AJ, Chenevert T, Collins D, Lehman C, Matos C, et al. Koh DM Diffusion-weighted imaging outside the brain: Consensus statement from an ISMRM-sponsored workshop. J Magn Reson Imaging 2016;44:521–40. https:// doi.org/10.1002/jmri.25196. Epub 2016 Feb 19. [22] Driessen JP, Caldas-Magalhaes J, Janssen LM, et al. Diffusion-weighted MR imaging in laryngeal and hypopharyngeal carcinoma: association between apparent diffusion coefficient and histologic findings. Radiology 2014;272:456–63.
Please cite this article in press as: Preda L et al. MRI evaluation of sacral chordoma treated with carbon ion radiotherapy alone. Radiother Oncol (2017), https://doi.org/10.1016/j.radonc.2017.11.029