The current study was conducted to evaluate the long-term results of irradiation with carbon ions in a raster scanning technique in patients with skull base chordomas.
The current study was conducted to evaluate the long-term results of irradiation with carbon ions in a raster scanning technique in patients with skull base chordomas.
Between 1998 and 2008, a total of 155 patients (76 men and 79 women) with a median age of 48 years (range, 15 years-85 years) were irradiated with carbon ions using a raster scan technique. The irradiation was performed at the Society for Heavy Ion Research in Darmstadt, Germany. The median total dose was 60 gray (relative biological effectiveness) at 3 gray (relative biological effectiveness) per fraction. The median boost planning target volume was 70 mL (range, 2 mL-294 mL). Local control (LC) and overall survival (OS) were evaluated using the Kaplan-Meier method, whereas long-term toxicity was evaluated via questionnaires.
The median follow-up was 72 months (range, 12 months-165 months). All patients had residual macroscopic tumors at the initiation of radiotherapy. The authors observed 55 local recurrences during follow-up, as well as systemic disease progression in 4 patients. The resulting 3-year, 5-year, and 10-year LC rates were 82%, 72%, and 54%, respectively, whereas the 3-year, 5-year, and 10-year OS rates were 95%, 85%, and 75%, respectively. Age <48 years and a boost volume >75 mL were associated with a significantly improved LC and OS. Primary treatment resulted in a significantly better OS probability. No higher late toxicity could be detected after carbon ion treatment.
Carbon ion therapy appears to be a safe and effective treatment for patients with skull base chordoma, resulting in high LC and OS rates. Cancer 2014;120:3410–3417. © 2014 American Cancer Society.
Chordoma is a rare malignant bone tumor that arises from notochordal remnants. Therefore, chordomas are mostly located alongside the neuroaxis. The overall incidence of chordoma is 8.4 per 10 million, according to the Surveillance, Epidemiology, and End Results (SEER) database. To the best of our knowledge, no risk factors have been identified to date. Chordomas are usually low-grade tumors that grow locally and aggressively, with a very high recurrence rate.[2, 3] Chordomas are pathologically classified into classic, chondroid, and dedifferentiated subtypes, with the dedifferentiated subtype being associated with the worst prognosis, whereas the chondroid subtype has the best prognosis.[5-7] Despite a low potential for metastasis,[8, 9] local control (LC) is the most important prognostic factor for survival. Complete surgical removal as the cornerstone of treatment is often difficult to achieve with acceptable functional outcome due to a local invasive growth pattern with directly adjacent structures of the skull base. Hence, there is often macroscopic residual tumor after surgery, which results in insufficient tumor control with a worse overall outcome. Therefore, radiotherapy is required as an adjuvant or additive therapy after a worthwhile function-conserving tumor resection. A dose escalation up to at least 70 gray (Gy) (relative biological effectiveness [RBE]) appears to be necessary for sufficient LC. To apply such high doses, high conformal techniques such as proton or carbon ion beam therapy appear to be favorable due to the critical structures next to the irradiated tumor. In addition to these reasons for the use of protons, carbon ions seem to have a higher RBE compared with photons or protons. In a study regarding the use of carbon ions for the treatment of patients with skull base chordoma, our department found excellent tumor control rates of 70% at 5 years. The purpose of the current retrospective analysis was to evaluate long-term results in all patients with skull base chordomas who were treated with carbon ions using an active raster scanning technique at the Society for Heavy Ion Research (GSI) in Darmstadt, Germany.
All patients with chordoma of the skull base who had been treated with carbon ions between 1998 and 2008 at the GSI and who had a follow-up of at least 12 months were included in the current analysis. Patients with previous radiotherapy were excluded.
Patients with histologically proven skull base chordoma were immobilized using a precision scotchcast head mask that ensured a reposition accuracy of 1 mm to 2 mm. A stereotactic method was used to localize the target point. For 3-dimensional treatment planning, a computed tomography (3-mm slices) examination was required in all patients. The contouring of treatment volume and organs at risk was performed after 3-dimensional image correlation with T2-weighted and contrast-enhanced T1-weighted magnetic resonance imaging (MRI). The boost planning target volume (PTV2) includes the macroscopic (gross) tumor volume (GTV) visible on the MRI with a 2-mm safety margin. The initial PTV (PTV1) includes the GTV and suspected subclinical disease. The treatment was performed at the GSI in Darmstadt using carbon ions in a raster scanning technique. Further detailed information regarding the treatment planning and carbon ion treatment has been published previously.
After approval from the ethics committee at the University of Heidelberg, all patients received a follow-up information sheet regarding the current retrospective analysis and questionnaires concerning their side effects and tumor status. After the patient provided signed informed consent, we in addition evaluated all medical records and contacted other treating physicians, if required. The follow-up MRI scans were evaluated for all patients to determine the tumor status.
To evaluate the rates for LC and overall survival (OS), the Kaplan-Meier method was used. Age, tumor volume, sex, dose, and time of treatment were tested for their prognostic significance using the log-rank test. Long-term toxicity was measured by specifications in medical records and questionnaires. Due to the retrospective analysis, the side effects could only be noted quantitatively, as previously prescribed. The 10-year follow-up was divided into 6 periods: baseline, 0 to 1 years, 1 to 3 years, 3 to 5 years, 5 to 7 years, and 7 to 10 years. For all endpoint curves, the unadjusted and adjusted collective were evaluated. In the unadjusted collective, the side effects of the last contact were fixed and transferred to the following time intervals. Therefore, the values always refer to the total cohort of all 155 patients. In the adjusted collective, patients dropped out after their last contact. Thus, the counted values of side effects in each period refer to the examined patient number during each period.
A total of 155 patients with histologically confirmed skull base chordoma were included and analyzed. The collective included 76 males and 79 females with a median age of 46 years at the time of first diagnosis (range, 13 years-85 years) and 48 years (range, 15 years-85 years) at the time of the initiation of carbon ion treatment. A total of 101 patients were irradiated after undergoing biopsy (16 patients) or R2 surgical resection (85 patients), whereas 54 patients were treated for recurrent disease after undergoing R2 surgical resection in the past. Hence, all 155 patients had macroscopic tumor at the time of treatment initiation. No patient received systemic therapy before or during radiotherapy. Sufficient data for PTV2 analysis was available for 93% of the patients. The median boost volume was 70 mL (range, 2 mL-294 mL). The delivered dose was a median of 60 Gy (RBE) (Table 1). The total dose was applied in a median weekly fractionation of 7×3 Gy (RBE) carbon ions. Consequently, the corresponding median equivalent dose calculated for an alpha/beta of 2 Gy and a fraction dose of 2 Gy (ED2Gy) to the chordoma was 75 Gy (RBE). The median follow-up after treatment was 72 months (range, 12-165 months).
|Dose, C12 in Gy RBE|
|Boost volume, mLa|
A total of 55 patients (35%) developed a local recurrence within the follow-up period of the study. Approximately 76% (42 patients) of these local recurrences occurred within the boost volume (PTV2), 16% (9 patients) occurred within the PTV1, and 7% (4 patients) occurred outside of the treatment field. Four patients developed metastatic disease. The 3-year, 5-year, and 10-year LC probability was 82%, 72%, and 54%, respectively (Fig. 1). A significantly higher LC rate could be identified in patients aged <48 years compared with patients aged ≥48 years (P = .033). Patients aged <30 years had an LC probability of 86% after 10 years compared with 47% in patients aged ≥30 years (P = .009). A PTV2 <75 mL was associated with a significantly (P = .002) better LC rate than a volume >75 mL (Fig. 2). A trend toward higher LC values was evident in female patients, in patients who received doses >60 Gy (RBE), and in patients after primary treatment (Table 2). The resulting 3-year, 5-year, and 10-year OS probabilities after carbon ion treatment were 95%, 85%, and 75%, respectively (Fig. 1). Younger age and a small PTV2 <75 mL were positive prognostic factors for OS. The 10-year OS probability in patients aged <48 years was 92% and was 58% in patients aged ≥48 years (P<.001). Furthermore, no patient aged <30 years died during the follow-up. A boost volume <75 mL resulted in a 10-year OS rate of 84%, compared with 58% with a boost volume >75 mL (P = .030) (Fig. 2). Primary treatment resulted in a significantly better 10-year OS probability compared with after recurrent treatment (83% vs 60%; P = .025) (Table 2). In addition, the OS probability from first diagnosis was evaluated. The median follow-up was 88 months with a respective 3-year, 5-year, and 10-year OS probability of 99%, 94%, and 80% after first diagnosis. Table 2 shows detailed results of the subgroup analyses.
|Factor||5-Year LC/OS||10-Year LC/OS||P|
|All patients||72% (LC)||54% (LC)|
|85% (OS)||75% (OS)|
|Sex||76% % vs 67% (LC)||59% vs 49% (LC)||NS|
|Female vs male||82% vs 87% (OS)||74% vs 75% (OS)||NS|
|Boost volume, mL||54% vs 84% (LC)||41% vs 62% (LC)||.002|
|>75 vs <75||81% vs 87% (OS)||58% vs 84% (OS)||.030|
|Dose, Gy RBE||88% vs 69% (LC)||69% vs 52% (LC)||NS|
|>60 vs ≤60||88% vs 84% (OS)||78% vs 74% (OS)||NS|
|Age, y||65 % vs 78% (LC)||42% vs 64% (LC)||.033|
|≥48 vs <48||76% vs 94% (OS)||58% vs 92% (OS)||<.001|
|Age, y||67% vs 92% (LC)||47% vs 86% (LC)||.009|
|≥30 vs <30||82% vs 100% (OS)||70% vs 100% (OS)||.008|
|Treatment||76% vs 64% (LC)||62% vs 40% (LC)||NS|
|Primary vs recurrent||86% vs 81% (OS)||83% vs 60% (OS)||.025|
As described earlier, the 10-year follow-up was divided into 6 periods to quantitatively evaluate the side effects after carbon ion treatment. The evaluated patient numbers in each period were 155 patients at baseline, 128 patients at 0 to 1 year, 109 patients at 1 to 3 years, 99 patients at 3 to 5 years, 73 patients at 5 to 7 years, and 51 patients at 7 to 10 years. Mucositis, xerostomia, dysgeusia, or alopecia occurred as acute side effects after irradiation in approximately 15% of the patients, before decreasing to initial values within the first year after treatment. Hearing deficits (17% at baseline) increased to 29% unadjusted and 34% adjusted, respectively, in the first year after treatment. However, the hearing deficits decreased during the further follow-up. After 5 years, the values were 14% (unadjusted) and 14% (adjusted), which was comparable to those at baseline. The most frequently mentioned symptom at baseline was double vision, as reported by 45% of all patients. After treatment with carbon ions, reports of this symptom decreased to 28% (unadjusted) and 26% (adjusted). Dizziness (22% at baseline) slightly increased over time, up to 32% (unadjusted) and 29% (adjusted). After an initial peak at 12% (unadjusted) and 14% (adjusted), fatigue symptoms (8% at baseline) demonstrated a marked decrease over time. There was a slight increase in the number of patients with seizures between the baseline (2%) and 7 years to 10 years of follow-up (7% [unadjusted] and 4% [adjusted]). In general, 62% of the patients reported cranial nerve deficits at the initiation of treatment. After 1 to 3 years, there was a reduction of approximately 20% (unadjusted) and 25% (adjusted) in cranial nerve palsies. After 7 years to 10 years of follow-up, 54% (unadjusted) and 53% (adjusted) of the patients again reported cranial nerve deficits. All side effects are listed in Table 3. No treatment-related secondary malignancies were observed during follow-up.
|Baseline (n=155)||Year 0 to 1 (n=128)||Year 1 to 3 (n=109)||Year 3 to 5 (n=99)||Year 5 to 7 (n=73)||Year 7 to 10 (n=51)|
|Double vision, %||45||36||34||28||25||25||23||27||30||28||26|
|Visual field deficits, %||13||12||12||12||11||14||16||16||8||12||4|
|Hearing deficits, %||17||29||34||25||27||19||18||14||14||16||22|
|Cranial nerve deficits, %||62||48||45||39||35||43||43||48||49||54||53|
Due to the rareness of chordoma, to the best of our knowledge the only published results are retrospective analyses taken from small groups of patients. With 155 patients, the current study is to our knowledge the largest group examined after carbon ion treatment and the second largest group of patients with skull base chordoma after radiotherapy published to date. The Boston group published data from 169 patients with a median follow-up of 41 months. At 72 months, the median follow-up in the current study was approximately 3 years longer. Their 5-year and 10-year LC rates were 73% and 54%, respectively. The corresponding OS rates were 80% at 5 years and 54% at 10 years after treatment with the photon-proton combination. Results of the Centre de Protontherapie d'Orsay demonstrated respective 4-year LC and OS rates of 54% and 80% after radiotherapy with a combination of protons and photons in 100 patients with clival chordoma. The median follow-up in this group was 31 months. As a supplement, the published results of the Loma Linda University Medical Center group, the Paul Scherer Institut (PSI), and the National Institute of Radiological Science (NIRS) are listed in Table 4. The evaluation of all published results regarding particle therapy demonstrates superior results regarding LC and survival rates in comparison with conventional techniques. In a meta-analysis of Amichetti et al, an average 5-year LC rate of 69% and an average 5-year OS rate of 80% were found after proton therapy. In patients with chordoma, a dose escalation up to at least 70 Gy (RBE) is necessary to achieve sufficient tumor control. Therefore, high conformal techniques are needed to apply a sufficient dose while sparing the organs at risk. Protons or carbon ions alone or in combination with photons are currently the international standard in the radiotherapy for patients with skull base chordoma. In addition to the benefits of proton therapy, carbon ions appear to have a higher RBE in the target region. The 10-year results of our evaluation confirm the effectiveness of carbon ion treatment. Furthermore, the clinical effect of the higher RBE of carbon ions is currently being evaluated in a randomized phase 3 trial comparing proton versus carbon ion therapy in patients with chordomas of the skull base. To the best of our knowledge, there are no randomized data with which to compare particle therapy with photon therapy, which will most likely always be the case due to the aforementioned advantages of particle therapy. Although a 10-year LC rate of 54% is a very good outcome for patients with chordoma, there is still room for improvement. Because 76% of the local recurrences in the current study occurred in PTV2, a further dose escalation may possibly lead to better results. Therefore, we currently treat patients with a slightly increased dose in our randomized phase 3 trial. A combination of radiotherapy and systemic therapy or neoadjuvant carbon ion treatment are other possible approaches to increase the effectiveness of the treatment in the future.
|Study||Radiation||TD, Gy RBE||Dpf, Gy RBE||No.||Follow-Up, Months||LC||OS|
|Hug 199920||P+Ph||71.9 (66.6-79.2)||1.8||33||33.2||3-y: 67%||3-y: 87%|
|5-y: 59%||5-y: 79%|
|Munzenrider 199918||P+Ph||66-83||1.8-1.92||169||41||5-y: 73%||5-y: 80%|
|10-y: 54%||10-y: 54%|
|Noel 200519||P+Ph||67 (60-71)||1.8-2.0||100||31||2-y: 86%||2-y: 94%|
|4-y: 53%||4-y: 90%|
|Ares 200921||P||73.5 (67-74)||1.8-2.0||42||38||3-y: 87%||5-y: 62%|
|Mizoe 200922||C||48-60.8 (4 wk)||3-3.8||33||53 (mean)||5-y: 85%||5-y: 88%|
|10-y: 64%||10-y: 67%|
|Current study||C||60 (57-70) (3 wk)||3-3.5||155||72||3-y: 82%||3-y: 95%|
|5-y: 72%||5-y: 85%|
|10-y: 54%||10-y: 75%|
In the current study, we found a strong significant favor in LC and OS for younger patients (those aged <48 years). Furthermore, the 10-year OS rate for patients aged <30 years was 100%, with a 10-year LC rate of 86%. The data published by some proton centers have presented similar results in children compared with adults, with 5-year LC rates of 60% to 81% and 5-year OS rates of 60% to 89%.[24-27] At the Massachusetts General Hospital in Boston, the OS rate in 73 children with skull base chordoma was 81% after a mean follow-up of 7.25 years. The 25 patients in the current study who were aged <30 years were found to have superior 5-year LC and OS results of 91% and 100%, respectively. Noel et al also found that patient age <52 years correlated with a better outcome. However, it remains unclear whether younger patients present with a less aggressive histology or if clinical symptoms lead to an early diagnosis and thus to the earlier initiation of therapy.
In several proton centers, the amount of residual tumor volume at the time of treatment initiation has been found to be of prognostic value. At Loma Linda University Medical Center, 100% of patients with chordoma and chondrosarcoma with a GTV <25 mL were still alive and with LC 5 years after treatment. Although the cutoff GTV of 25 mL was confirmed by the PSI, again only a combined evaluation of patients with chordoma and chondrosarcoma was performed. These results are supported by the data from the current study demonstrating a highly significant advantage in terms of LC and OS in patients with a PTV2 (boost volume) <75 mL compared with those with a PTV2 of >75 mL.
Sex was not found to be a significant prognostic factor in the current study and its role remains unclear because there are inconsistent published results. In the PSI series, sex was not found to be statistically significant21; however, in the Massachusetts General Hospital series, a better 5-year and 10-year LC rate for male patients (81% and 65%, respectively) versus female patients (65% and 42%, respectively) was found.
To our knowledge, there still is no consensus regarding whether radiotherapy should be performed immediately after surgery for clival chordoma or held until disease recurrence occurs. However, there is some evidence that immediate irradiation leads to better results in the treatment of patients with sacral chordoma. In the current study, 61 patients were treated in a recurrent situation after having undergone R2 surgical resection in the past. We found a significantly (P = .025) higher 10-year OS rate in patients after primary treatment (83%) compared with those after recurrent treatment (60%). The corresponding 5-year and 10-year LC rates were 76% and 62%, respectively, versus 64% and 39%, respectively, suggesting that early additive treatment appears to be beneficial (P = .063).
Radiotherapy with a total dose >60 Gy (RBE) led to a nonsignificant trend toward better control rates compared with an irradiation dose of ≤60 Gy (RBE). One reason for the nonsignificance could be the small number of patients (17 patients) who received doses >60 Gy (RBE). Schulz-Ertner et al postulated a dose response relationship with a further improvement in LC with doses >75 Gy (RBE) (at a dose of 2 Gy [RBE] per fraction) based on a meta-analysis. Hence, an adaption to higher total doses was conducted in our prospective randomized phase 3 trial, which is currently recruiting patients.
Through retrospective analysis, we quantitatively evaluated the side effects after carbon ion treatment. As has been previously reported, carbon ion treatment in patients with clival chordoma led to mild acute toxicity.
Hearing problems increased within the first year after treatment, but returned to the baseline values during the subsequent follow-up. This course can be explained by the appearance of middle ear effusion after irradiation, which is treatable with a tympanostomy tube. In our long-term evaluation, no patient developed high-grade toxicity regarding the inner ear and Nervus vestibulocochlearis.
Hasegawa et al pointed out a correlation between visual loss and carbon ion treatment with doses >60 Gy (RBE) to 20% of the optic nerve volume. Munzenrider et al reported that approximately 4% of their treated patients experienced visual function problems after proton therapy with a median dose application of 62.1 Gy (RBE) to their optic structures. The data from the current study demonstrated a relatively constant number of patients with visual field limitations at baseline and after 10 years. Approximately 2% of patients (3 patients) reported a decreased visual field after treatment. The dose to the optic structures in the patients in the current study was restricted to 54 Gy (RBE).
Urie et al reported cranial nerve injuries at a rate of 1% at 60 Gy (RBE) and 5% at 70 Gy (RBE) after proton therapy. Greater than 50% of the injuries occurred within the first 24 months after treatment, but none occurred after 60 months. In the current evaluation, after an initial decrease in the rate of cranial nerve deficits at year 1 to 3, an increase of approximately 15% (unadjusted) to 18% (adjusted) was again observable. After the evaluation of every patient over the entire observation period, no change in cranial nerve status was found in 59 patients, whereas 20 patients demonstrated a local recurrence. A total of 47 patients demonstrated an improvement in their cranial nerve deficits, although 17 of these patients developed a local recurrence. An additional 19 patients demonstrated an impairment after irradiation, only 6 of whom were found to have recurrent disease. A group of 30 patients had, after initial improvement, again experienced increased cranial nerve deficits. Of these 30 patients, only 12 patients developed disease recurrence during the follow-up period. Thus, the recrudescence of cranial nerve deficits in the long-term follow-up appears to be caused on the one hand by those patients with recurrent chordoma and conversely by radiotherapy after initial improvement due to the tumor shrinkage. However, in summary, fewer patients reported cranial nerve palsies at 7 to 10 years after treatment compared with at the initiation of treatment.
Debus et al found an increasing risk of brain stem toxicity if ≥5.9 mL of tumor received ≥50 Gy (RBE), ≥2.7 mL received ≥55 Gy (RBE), or if ≥0.9 mL received ≥60 Gy (RBE). With a maximum dose to the brainstem of 50 Gy (RBE) at the center and 60 Gy (RBE) to the surface (<1 mL), we did not observe any toxicity to the brain stem in the patients in the current study after carbon ion treatment.
A lower application of dose to the normal tissue and a lower neutron exposure to the whole body by using particles is assumed, compared with photon therapy. The improvement is even higher with carbon ions and using a spot scanning delivery method. Nevertheless, the risk of secondary malignancies after carbon ion treatment remains subject to discussion. No treatment-related secondary malignancies were noted among the patients in the current study during the follow-up.
The use of retrospective analysis is certainly a limiting factor in the current study. However, to the best of our knowledge there are still no prospective trials published to date and the retrospective series usually involves a small number of patients with a short follow-up. Therefore, this evaluation provides unique and important results for patients and oncologists.
Carbon ion therapy is a safe and effective treatment among patients with chordoma of the skull base. Young age and small tumor volume are prognostic factors for higher LC and OS rates. Primary treatment resulted in higher OS rates compared with the treatment of patients with recurrent skull base chordoma after surgery. Based on our primary findings, we initiated a randomized phase 3 trial in 2009, comparing carbon ions with protons, which is still open for patient accrual.
Partially funded by grants from German Cancer Aid and the German Research Foundation (grant KFO 214).
Dr. Debus is Chief Executive Officer of the Heidelberg Ion Beam Therapy Center (HIT).