Skull Base Chordoma: Long-Term Observation and Evaluation of Prognostic Factors after Surgical Resection
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.333474
Source of Support: None, Conflict of Interest: None
Keywords: Karnofsky-Performance Scale, prognostic factors, radiotherapy, skull base chordoma
Chordoma is a rare bone malignancy accounting for 1–4% of the primary skeletal tumors and was first described microscopically by Virchow in 1857., It most commonly manifests itself at the distal and proximal ends of the spine. Skull base chordoma (SBC) accounts for 25–30% of all chordomas, which are believed to originate from embryonic remnants of the primitive notochord. The challenge in the treatment lies in the deep cranial localization, proximity to vital structures like the brain stem, and their high recurrence rate.,, The gross total surgical resection remains the treatment of choice. Radiotherapy as an adjuvant treatment with surgery can offer in selected cases a long-term tumor control, whereas the sensitivity to chemotherapy is very low.,, Prognostic factors, which have been linked to the outcome, include the extent of resection, prior surgical records, pathology, and radiotherapy.,, In the current study, we examined prognostic factors on the overall survival after surgical resection of SBC in this institution.
A retrospective review of 41 patients, who had undergone surgical resection for SBC at PD Hinduja National Hospital, Mumbai, India, by the senior author between January 2001 and November 2017 was undertaken. The inclusion criteria for the study were histological confirmation, documented imaging, and treatment of recurrent disease and where the histological report was known. In this context, medical charts and imaging were analyzed for the following data: sex, age, prior clinical history and status, evaluation of the Karnofsky-Performance Scale (KPS), surgical reports, radiological reports, and radiotherapy.
All procedures performed in the studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. For this type of study, formal consent is not required.
In view of the tumor location and pattern of invasion, different surgical approaches were taken into account in the decision-making process. Whenever possible, radical excision was chosen by the same experienced surgeon, always taking into consideration the preservation of neurological function while aiming for maximum safe cytoreductive surgery. A summary of the surgical approaches is comprised in [Table 1]. The tissue samples were reviewed by experienced pathologists classifying them in accordance with the accepted histological criteria. The expansive nature of the chordomas like dural invasion and encasement of the internal carotid artery (ICA) were documented through imaging and intraoperative assessment [Table 2].
Adjuvant radiotherapy with gamma-knife radiosurgery (GKRS) was delivered to 15 patients at our institution, who, after assessment, were concluded to have an inoperable residual tumor. Frequent follow-up examinations were conducted and updated magnetic resonance imaging (MRI) scans were done to monitor the progression of the disease. All patients from the cohort were called upon via telephone interview in October 2017 to update on their latest status. Eight patients could not be contacted.
The data were collected from medical charts and reports and compiled with Microsoft Excel 2012. Statistical analyses were conducted using SPSS 23 (Statistical Package for the Social Sciences, IBM). The surviving analyses were performed using the Kaplan–Meier procedure. The survival time was defined from the first diagnosis of chordoma until death. Data were censored at the time of an event (death) or at the end of the latest follow-up at 192 months (16 years) if the event did not occur. A comparison of the survival times between patients with and without radiotherapy was conducted using the log-rank test. Cox regression analysis was performed for investigating the effect of age, KPS, ICA encasement, and dural invasion upon survival time. The hazard ratio (HR) and 95% confidence interval (CI) were applicable. For correlation analysis between pre-surgical KPS, and at the last follow-up, Kendall' s-Tau-b correlation coefficient was used. For the correlation analysis between the number of surgeries and KPS at the last follow-up and receiving radiotherapy and KPS at the last follow-up, Pearson's correlation coefficient was used. The interval-scaled data are presented as mean and standard deviation and nominal data as frequency and valid percent. A P value of <0.05 was considered statistically significant.
From January 2001 to November 2017, 41 patients with primary and recurrent disease were admitted and treated at our institution [Table 2]. Thirty-four patients with the primary disease (first diagnosed) and 7 patients with recurrent skull base chordoma (SBC) (initially operated elsewhere) were operated on by a single experienced surgeon during the time period. The most common clinical symptoms were headache (63%) and problems of vision (54%) like diplopia. In the clinical examination, paresis of the abducens nerve (32%) and the lower cranial nerves (LCN) (27%) was most common. Preoperative MRI scans were done prior to surgery to decide on the treatment strategy.
The gold standard to treat SBC is surgical resection. In the case of recurrent disease, 51.2% of the cohort was operated on once, while 48.8% received more than one operation [Table 2]. About 36.6% of the patients were given adjuvant GKRS with a median prescription dose of 16 Gy and a dosage range between 13 and 18 Gy.
Factors affecting overall survival
The follow-up data were available on 41 cases and clinical evaluation took place for up to 192 months. The mean survival time for the patients was 123.37 months (95% CI 90.89–155.86) [Figure 1]. In the cohort of 41 patients, a total of 10 patients died due to tumor progression. The 5- and 10-year survival rates were 73.3 and 49%, respectively. There was no significant impact on the overall survival comparing genders (P = 0.606) and age (P = 0.708).
The average preoperative KPS was 81 ± 15.1 (mean ± SD). In the last follow-up examination, the KPS was 83 ± 17.9 (mean ± SD) for the surviving patients. The Cox regression analysis showed that there was a significant relationship between the survival rates in the overall study population and pre-surgery KPS (P = 0.004, HR 0.96, 95% CI 0.93–0.99). This demonstrated that the survival rate increased by 4.5% with each 10 points decrease of pre-surgery KPS. This was further supported with a positive significant correlation between the pre-surgery KPS and the KPS at the last follow-up (P = 0.039). Examining whether different treatment modalities like the number of surgeries versus surgery and postoperative radiotherapy had an impact on the last follow-up KPS showed that there was no correlation between the number of surgeries and the last follow-up KPS (P = 0.097). There was no correlation between whether the patients received radiotherapy or not and the last follow-up KPS (P = 0.431).
After 5 years, 69.2% of the patient with one-time surgery and 71.2% of the patients with more than one surgery survived. The mean survival time for the patients with one-time surgery was 73.85 months and 125.37 months for the patients with more than one surgery. A log-rank test was used to determine if there were differences in the survival functions between one-time surgery and repeated surgery. The survival functions of the two interventions were not significantly different (P = 0.961).
The mean survival time for the patients with radiotherapy was 152.0 months and for the patients without radiotherapy 106.42 months. After 5 years, 88.9% of the patients receiving radiotherapy and 64.1% of the patients that did not receive radiotherapy, were still alive. At 192 months, the survival rate for the patients with radiotherapy was 66.7% and for the patients without radiotherapy was 38.5%, but with no statistically significant differences between the survival functions of both groups (log-rank test, P = 0.117, [Figure 2]).
Impact of tumor infiltration
Due to the high rate of recurrence and invasion into vital structures, tumor infiltration on the overall survival and clinical condition was assessed as well. The parameters evaluated in this study were dural infiltration [Figure 3] and ICA encasement [Figure 4] assessed in preoperative MRI scans and intraoperatively. From the overall cohort of 41 patients, 8 patients showed dural invasion and 8 patients an encasement of the ICA.
The analyzed variables of ACI-encasement and dural invasion showed no significant effect on the overall survival in the Cox regression analysis. Regarding histopathology, 15% of the patients had chondroid chordoma and 85% conventional chordoma. There was no significant difference in the overall survival, however, when comparing chondroid and conventional chordoma, P = 0.554.
The current study supports headache and problems of vision like diplopia as frequent clinical symptoms in SBC. Clival chordoma often results in cranial nerve palsies. The most common clinical manifestations in our study were paresis of the abducens nerve and the lower cranial nerves, as mentioned by Kano. Nevertheless, the clinical manifestations vary and depend predominantly on the location.
Even though in our study there is a predominance of the lesion among men, this did not result in any effects on the overall outcome.,, Also age did not show any association with the outcome, thus, being consistent with the reports by Zhang and McMaster, who analyzed 400 chordoma cases from the SEER database (Surveillance, Epidemiology and End Results Program)., In a systematic review of 22 clinical case series with 1,754 patients, Zou et al. found out that preoperative visual deficits and elder patients were negative predictors for survival. Concerning the age of the patients, the studies show conflicting results. This could be due to pre-consisting illnesses of the patients, which increase with age.
The KPS has proven to be an important factor in the assessment of the overall survival, and thus, the therapeutic decision-making process. Meng et al. showed that a KPS ≥80 significantly improves the overall survival of the patients with spinal chordoma. Interestingly, in our study, higher pre-surgery KPS resulted in significantly higher survival rates (P = 0.004). Supporting this was also a positive significant correlation between the pre-surgery KPS and the KPS at the last follow-up independent of the treatment modalities (P = 0.039). As with most case series on chordoma, one has to take into account the limited number in the case pool of this study. The tumor pathology did not show any significant difference in the overall survival between chondroid and conventional chordomas. Similar findings were also reported by Colli, Forsyth, Boari, and Safwat.,,,
The advances in radiation technology have led to treating chordoma with proton-beam therapy., Fuji et al. and Hug et al. have shown that proton-beam therapy resulted in a better survival outcome. Kano and Yoneoka have reported that radiosurgery post-resection offers a good therapeutic outcome, especially for residual lesions., However, proton-beam therapy is not available at our institution, thus, GKRS was used for the treatment. GKRS has been found to be useful after gross tumor reduction. In this study, the mean survival time for patients with GKRS was 152.0 months and without 106.42 months, which indicates that the use of radiosurgery prolongs the mean survival time.
Being slow in growth and often clinically dormant until the late stages of the disease, SBC can be most challenging to operate for a neurosurgeon. Advancements in microsurgical techniques, as well as endoscopic endonasal techniques, have been well-documented in the literature., Due to the recurrent nature of the chordomas, the current treatment is repeat surgical resection. This was also mentioned by Al-Mefty, who advocated radical resection to aim for long disease-free survival. Therefore, often multiple approaches, as well as extradural approaches for radical reduction, should be taken into mind to prevent recurrence and spreading via the cerebrospinal fluid. This was taken into consideration in the different operative approaches in this study [Table 1]., The mean survival time for patients with one-time surgery was 73.85 months and 125.37 months for patients with more than one surgery. Even though the difference was not significant (P = 0.961), the results show that repeat surgeries increase the mean survival. Still, one has to keep the risk of morbidity associated with reoperation in mind, as mentioned by Walcott.
Another important factor to be taken into consideration is the tumor infiltrating into vital structures, thus, having an impact on the overall survival. Zhang has stated that adherence to vital structures could be a risk factor for survival. In the current study, tumor encasement of the ICA and dural invasion did not have any significant effect on the overall survival. One has to take into account that in this study, the population could have been too small to show a significant effect. The literature is conflicting on tumor infiltration as a prognostic factor. In a recent study by Zhai et al., they were able to show in a retrospective study of 67 patients that dural penetration is a prognostic factor for overall survival. Boari et al. has highlighted other prognostic factors as well. They were able to show that SBC invading the rhinopharynx had a significant impact on overall the outcome. However, the dural invasion did not have any significant impact on the outcome. Additionally, Wu et al. has reported that the subtotal resection and recurrent tumor were statistically significant predictors with worse long-term outcomes in 106 patients on univariant analysis. Thus, recurrence may cause infiltration of the tumor tissue into vital structures such as ICA encasement or dural invasion.
The currently conducted studies to find predictive clinical features for prognosis and outcome of chordoma are still inconclusive and controversial. Given that, clinical studies with larger cohorts are necessary. Hereby, molecular biomarkers may be more reliable and fundamental as prognostic factors and therapeutic decision-making.
SBCs are difficult to treat due to deep cranial localization and being clinically dormant until the late stages of the disease. The most extensive surgical resection with adjuvant radiotherapy and repeat surgery of recurrent disease are currently the treatment of choice. In our study, tumor encasement of the ICA and dural invasion did not have any significant effect on the overall survival. Statistical analysis showed that repeat surgical resection and radiotherapy could be prognostic factors. In this study, we were able to show that mortality decreased by 4.5% with each 10 point increase of pre-surgery KPS. This could be an additional decision-making factor for treatment modalities.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2]