Abstract
Background
Microsurgical resection after failed radiosurgery (SRS) in vestibular schwannoma (VS) patients is associated with higher morbidity. Identifying factors that predict treatment failure (TF) is crucial. Additionally, distinguishing between pseudoprogression (PP) and true tumor progression (TP) can be challenging. This study aims to identify predictive factors for TF and investigate early features that differentiate PP from TP.
Methods
A retrospective analysis was performed on 705 patients with unilateral sporadic VS who underwent SRS between 1998 and 2020. Clinical data, including patient characteristics, symptoms, tumor volume (TV), and onset of new symptoms after SRS, were recorded. The average follow-up was 4 years for the TP group (n = 107) and 7 years 10 months for the remission group (n = 598).
Results
TF was more common in women (P = .04) and linked to lower OHATA class (P = .03). Age, clinical symptoms, TV, and configuration (cystic vs solid) were not predictive of TF. TP-patients experienced significantly more new neurological symptoms (20.6% vs 8.4%, P < .001), especially hemifacial spasm (P < .001), which was associated with OHATA class (A > B > C > D > E, P = .001). Relative TVs (RTV) differed significantly between TP and tumor control (TC) groups, with the TP group showing higher RTV at both 12 months (TC = 1.0 ± 0.6, TP = 1.4 ± 1, P = .002) and even more at 24 months (TC = 0.71 ± 0.5, TP = 1.5 ± 0.7, P < .001) after SRS.
Conclusions
Female sex and lower OHATA class were identified as independent predictors of TF. Hemifacial spasm occurrence after SRS was linked to TP. Failure of reduction of initial TV (RTV > 1) after 24 months was associated with TP with a high sensitivity and specificity, making PP unlikely.
Keywords: hemifacial spasm, radioresistance, stereotactic radiosurgery, treatment failure, vestibular schwannoma
Key Points.
Female sex and OHATA class are risk factors for treatment failure after radiosurgery in VS.
The onset of HFS after SRS is associated with tumor progression.
In case of continuous tumor volume increase 24 months after treatment, TP is highly suspected.
Importance of the Study.
This study is critical in advancing our understanding of treatment outcomes in patients with vestibular schwannoma (VS) who undergo stereotactic radiosurgery (SRS). Although SRS is an effective treatment, treatment failure (TF) and the differentiation between pseudoprogression (PP) and tumor progression (TP) can complicate clinical decisions. The study identifies key predictive factors, including female sex and OHATA class, that may contribute to TF, which could guide clinicians in tailoring patient counseling. Additionally, the study highlights the importance of monitoring tumor volume trends over time as a key indicator of treatment success. Understanding these factors not only helps in predicting the likelihood of TF but also in distinguishing between PP and TP early on, improving the accuracy of diagnosis and subsequent treatment decisions.
Vestibular schwannomas (VS) are benign intracranial lesions arising from the vestibulocochlear nerves. They are the most common tumors found in the cerebellopontine angle and account for 8-10% of intracranial tumors.1 Treatment options for sporadic VS include “wait and scan,” microsurgical resection, and radiation therapy.2 Stereotactic radiosurgery (SRS) is a viable treatment option for VS with long-term tumor control (TC) rates of around 88%–91%3,4; however, some tumors may be resistant to radiation. About 9%–12% of pre-irradiated VS show tumor progression (TP).5,6 The factors that determine radioresistance are not yet fully understood. Dysregulation of genes such as PTEN and mTOR, which are involved in pathways that are known to be altered upon irradiation, may play a role in radioresistance in NF2-associated VS.6 In addition, other mechanisms determining tumor hypoxia, mutations affecting tumor suppressors and oncogenes, activation of DNA repair mechanisms, prolonged cell cycle arrest, aberrant expression of cell cycle checkpoint proteins, and/or radiation dose may contribute to radiation resistance.5 Understanding the radiobiology of VS is therefore fundamental to optimize the current treatment strategies. However, personalized treatment based on molecular alterations is only possible after analyzing the tumor tissue at a molecular level. An earlier identification of patient- and tumor-related factors that influence radioresistance could help in clinical decision making to predict which cases are more likely to experience treatment failure (TF) after SRS.
Another important and common aspect is a transient increase in VS volume after SRS, defined as pseudoprogression (PP). This complicates the differentiation between treatment-related changes (PP) and TF.3,7 The appropriate duration for post-treatment surveillance is still a topic of ongoing debate. To date, there are no clear criteria that define this phenomenon in terms of the percentage of volume increase or the time course observed.
The main aim of this study was to identify factors that could predict TP after Gamma-Knife SRS. Therefore, a large cohort of patients treated at the same high-volume Gamma Knife center with long-term follow-up were investigated to determine patient and tumor-related factors that may influence TF. The second objective was to investigate potential features that allow differentiating at an early-stage PP and TP by analyzing serial volumetric data.
Methods
Study Design and Patient Cohort
This retrospective cohort study analyzed a consecutive patient series with unilateral sporadic VS treated at one Gamma Knife center between December 1998 and February 2020. Among them, we selected all patients who developed TP following Gamma Knife treatment and compared them with the remaining population without TP. Patients with a follow-up <2 years, NF2-associated VS, or previously treated VS were excluded.
All patients received Gamma-Knife SRS with a prescription dose of 12–13 Gy to the 55%–65% isodose line. Follow-up imaging was scheduled at 6 and 12 months after SRS and annually thereafter.
Data Collection
Tumor size was graded according to the Hannover Classification system.8 The OHATA Classification was used to classify the lateral extent of the tumor in the internal auditory meatus,9 in particular:
- Type A: The tumor extends laterally to the fundus and involves the bony foramen.
- Type B: The tumor reaches the fundus but does not involve the bony foramen.
- Type C: The tumor is confined to the lateral half of the IAC, leaving a tumor-free space near the fundus.
- Type D: The tumor is located within the medial half of the IAC.
- Type E: The tumor is at the porus acusticus internus without causing enlargement of the IAC.
The clinical state of facial nerve and hearing function was reported by House and Brackmann (H&B)10 and Gardner-Robertson (G&R) scale.11 Secondary VS symptoms, such as tinnitus and vertigo, were measured with a 4-point scale (0 = none, 1 = occasionally, 2 = constantly, 3 = invalidating). Trigeminal symptoms were also recorded with a 4-point scale (0 = no symptoms, 1 = occasionally, 2 = hypesthesia, 3 = trigeminal neuralgia). An increase of one point was interpreted as a worsening of symptoms, while a decrease of one point was interpreted as an improvement. The tumor configuration (cystic vs solid) and the onset of new neurological symptoms after SRS were also recorded. The following radiation parameters were considered in this study: target volume (TaV—volume of the tumor in cm³), prescription isodose volume (PIV—volume of tissue in cm³ receiving the prescribed radiation dose), mean marginal dose in Gy (= J/kg), isodose (% used to define dose distribution), maximum dose (Dmax—the highest dose within the treatment area, typically corresponding to the 100% isodose line), absorbed dose in the TaV (mJ), and D90 (dose received by the 90% of the target volume).
Tumor increase was defined as an enlargement of tumor volume (TV) of more than 10%, while tumor reduction was defined as a shrinking of TV by more than 10%. TV was measured using slice-by-slice manual contouring by board-certified Gamma-Knife-Radiosurgery experts.
TP was defined as follows:
1) Progressive increase of TV on contrast-enhanced MR imaging after an observational period of at least 4 years.
2) New tumor growth after an initial decrease in TV.
3) An increase in TV causing new neurological symptoms or compression of important structures (even if follow-up was <4 years).
All other volume changes were classified as “pseudoprogression.”
Statistical Analysis
Statistical analysis was performed using IBM® SPSS Statistics 30 (IBM Corporation, Armonk, NY, USA). We used the Shapiro–Wilk test to analyze metric variables for normal distribution and compared data by T-tests or Mann–Whitney U test as appropriate. Nominally scaled data was analyzed employing the Chi-square test. Predictive factors identified as statistically significant in a univariable analysis were entered into a binary logistic regression model using a log-rank test. Kaplan–Meier curves were calculated to analyze parameters over time, and ROC curves were created to identify a cutoff for TP on follow-up. P-values <.05 were considered significant (confidence interval 95%).
Results
Study Cohort and Demographics
Among 705 patients with sporadic VS treated with SRS within a 22-year period, 107 individuals (15.2%) experienced a TP. In the TP group, n = 70 were female (65.4%) and n = 35 male (34.6%) patients with a mean age of 58 ± 11.4 years (range 33–79). The TC group was composed of n = 598 patients, n = 328 female (55%) and n = 270 male (45%) with a mean age of 58 ± 12.4 years. When comparing the 2 groups, it was found that females had a statistically higher incidence of TP (χ2 = 4.126, P = .04), while patient age was not associated with TP. The tumor was left-sided in 51.4% (n = 55) and 53.4% (n = 319), in the TP group and in the TC group, respectively, with no significant difference between the 2 groups.
The mean age of female patients in the TP group (57 ± 11.3 years) and TC group (58 ± 12.5 years) was comparable.
In the TP group, 28 patients (26.2%) underwent surgical tumor resection, while 78 patients (72.9%) had been re-irradiated. Only one 80-year-old patient from the TP group was not subjected to further treatment. In this specific case, following an initial phase of PP and subsequent reduction in TV, a second phase of tumor growth was observed. However, given the patient’s age and stable clinical condition, no further treatment was initiated.
Tumor Characteristics
Overall, 122 T1, 281 T2, 238 T3, and 64 T4 VS have been treated with SRS (Table 1). Tumor size at diagnosis in the TP group was most frequently grade T2 (39.9%), followed by T3a (19.9%), T1 (17.3%), T3b (13.9%), and T4a (7.9%). A similar distribution of TVs was observed in both the TP and the TC group (Figure 1A). The mean TV was 1.29 ± 1.5 cm3 and 1.51 ± 1.96 cm3 in the TP and the TC group, respectively. Based on this data, no linear association was observed between TV (U = 32422.5, P = .83) or size classified by Hannover grade (ꭓ2 = 5.563, P = .14) and TF. However, T1 tumors showed a significantly lower rate of TP than T2–T4 tumors.
Table 1.
Patient Demographics and Tumor Characteristics
| All N = 705 |
Tumor progression N = 107 |
Tumor control N = 598 |
P-value | |
|---|---|---|---|---|
| Age | 58 (±11.4) | 58 (±12.4) | ||
| Female | 398 (56.4%) | 70 (9.9%) | 328 (46.5%) | .042 |
| Male | 306 (43.4%) | 37 (5.2%) | 269 (38.2%) | |
| Tumor size (Hannover) | .218 | |||
| T1 | 122 (17.3%) | 11 (1.6%) | 111 (15.7%) | |
| T2 | 281 (39.9%) | 50 (7.1%) | 231 (32.8%) | |
| T3a | 140 (19.9%) | 24 (3.4%) | 116 (16.5%) | |
| T3b | 98 (13.9%) | 14 (2%) | 84 (11.9%) | |
| T4a | 56 (7.9%) | 8 (1.1%) | 48 (6.8%) | |
| T4b | 8 (1.1%) | 0 | 8 (1.1%) | |
| Tumor side | .711 | |||
| Right | 331 (46.9%) | 52 (7.3%) | 279 (39.6%) | |
| Left | 374 (53.1%) | 55 (7.8%) | 319 (45.3%) | |
| Cystic configuration | 24/461 (5.2%) | 9 (2%) | 15 (3.2%) | .089 |
| OHATA class | 558/705 | .064 | ||
| Class A | 116 (20.8%) | 18 (3.2%) | 98 (17.6%) | |
| Class B | 239 (42.8%) | 44 (7.9%) | 195 (34.9%) | |
| Class C | 157 (28.1%) | 14 (2.5%) | 143 (25.6%) | |
| Class D | 38 (6.8%) | 4 (0.7%) | 34 (6.1%) | |
| Class E | 8 (1.4%) | 0 | 8 (1.4%) |
P-values indicate significant differences. Values are presented as the number of patients (%) and mean ± SD, unless indicated otherwise. Significant P-values (<.05) are highlighted in bold.
Figure 1.
(A) TV targeted by SRS in the TC group (no progression) and the TP group (progression). No significant difference was observed. (B) TC according to the OHATA classification. TP was more frequent in classes A&B compared to classes C and D. No case of TP occurred in class E.
Additionally, OHATA class revealed a significant linear association with TP (ꭓ2 = 4.632, P = .03, see Figure 1B) as well as with the incidence of hemifacial spasm (HFS) after SRS (A[= 12.9%] > B[= 7.1%] > C[= 3.2%] > D[= 2.6%] > E[= 0.0%], ꭓ2 = 10.563, P = .001).
Tumor configuration was evaluated in 461 cases. Most tumors were solid (94.8%) in this cohort. Although cystic tumors were more frequent in the TP group (8.4%) in comparison with the TC group (4.2%), the difference was not statistically significant (ꭓ2 = 2.901, P = .089). However, TP occurred significantly earlier in patients with cystic tumors than in solid schwannomas (U = 652.5, P = .02, see Figure 2). It is noteworthy that recurrence rates were higher in both the solid and cystic tumor groups compared to the overall cohort, with rates of 22.4% and 37.5%, respectively. This is probably due to incomplete records over the 22-year study period, resulting in a significant number of missing cases (N = 244).
Figure 2.
Kaplan–Meier curve of progression-free survival in patients with solid and cystic VS. The overall incidence of TP did not differ between the 2 groups. However, cystic tumors tended to relapse earlier in comparison with solid schwannomas.
The data about patient demographics and tumor characteristics are summarized in Table 1.
Clinical Status and Outcome
Hypoacusis was the most common initial symptom, occurring in 78% of the TP group and 82% of the control group. Most TP-patients still had a functional hearing level (G&R 1 or 2) at diagnosis (57.9%), while 34.6% had already a non-serviceable (GR3) and 7.5% a poor hearing level (GR 4 or 5). Tinnitus occurred in 78.7% of cases in the TP group and in 75.2% of the TC group, while vertigo occurred in 60.8% and 60.5%, respectively.
In both groups, almost all patients had a fully intact facial nerve function (HB 1; 99% in the TP group, n = 106, 97.5% in the control group, n = 583). Trigeminal affection was rare, with 4.7% and 7%, respectively. Based on our data, the symptoms mentioned above (hypoacusis, tinnitus, vertigo, and facial and trigeminal nerve function) are not useful in predicting the risk of TF at diagnosis. The data regarding the clinical status at diagnosis and during follow-up are presented in Table 2.
Table 2.
Clinical Status at Diagnosis and During Follow-up
| All N = 705 |
Tumor progression N = 107 |
Tumor control N = 598 |
P-value | |
|---|---|---|---|---|
| Clinical status at diagnosis | ||||
| Hearing status (G&R) | 704/705 | .114 | ||
| 1 | 131 (18.6%) | 24 (3.4%) | 107 (15.2%) | |
| 2 | 228 (32.4%) | 38 (5.4%) | 190 (27%) | |
| 3 | 254 (36.1%) | 37 (5.3%) | 217 (30.8%) | |
| 4 | 58 (8.2%) | 8 (1.1%) | 50 (7.1%) | |
| 5 | 33 (4.7%) | 0 | 33 (4.7%) | |
| Tinnitus | 705/705 | .090 | ||
| None | 174 (24.7%) | 26 (3.7%) | 148 (21%) | |
| Occasionally | 250 (35.5%) | 40 (5.7%) | 210 (29.8%) | |
| Permanent | 262 (37.1%) | 34 (4.8%) | 228 (32.3%) | |
| Disabling | 19 (2.7%) | 7 (1%) | 12 (1.7%) | |
| Facial function (H&B) | 703/705 | .891 | ||
| 1 | 689 (98.1%) | 106 (15.1%) | 583 (83%) | |
| 2 | 4 (0.6%) | 1 (0.9%) | 3 (0.4%) | |
| 3 | 5 (0.7%) | 0 | 5 (0.7%) | |
| 4 | 3 (0.4%) | 0 | 3 (0.4%) | |
| 5 | 1 (0.1%) | 0 | 1(0.1%) | |
| 6 | 1 (0.1%) | 0 | 1(0.1%) | |
| Trigeminal nerve symptoms | 705/705 | .707 | ||
| None | 655 (92.9%) | 102 (14.5%) | 553 (78.4%) | |
| Occasionally | 12 (1.7%) | 2 (0.3%) | 10 (0.1%) | |
| Hypesthesia | 30 (4.3%) | 3 (0.4%) | 27 (3.8%) | |
| Pain | 8 (1.1%) | 0 | 8 (1.1%) | |
| Vertigo | 705/705 | .812 | ||
| None | 278 (39.4%) | 42 | 236 | |
| Occasionally | 284 (40.3%) | 45 | 239 | |
| Permanent | 128 (18.2%) | 19 | 109 | |
| Disabling | 15 (2.1%) | 1 | 14 | |
| Clinical status during FU | ||||
| Hearing Status (G&R) | 693/705 | .228 | ||
| 1 | 61 (8.8%) | 7 (1%) | 54 (7.8%) | |
| 2 | 128 (18.5%) | 17 (2.5%) | 111 (16%) | |
| 3 | 302 (43.6%) | 54 (7.8%) | 248 (35.8%) | |
| 4 | 139 (20.1%) | 24 (3.5%) | 115 (16.6%) | |
| 5 | 63 (9.1%) | 5 (0.7%) | 58 (8.4%) | |
| Tinnitus | 694/705 | .609 | ||
| None | 191 (27.5%) | 25 (3.6%) | 166 (23.9%) | |
| Occasionally | 239 (34.4%) | 42 (6.1%) | 197 (28.3%) | |
| Permanent | 248 (35.7%) | 37 (5.3%) | 211 (30.4%) | |
| Disabling | 16 (2.3%) | 3 (0.4%) | 13 (1.9%) | |
| Facial function (H&B) | 694/705 | .903 | ||
| 1 | 675 (97.3%) | 106 (15.3%) | 569 (82%) | |
| 2 | 9 (1.3%) | 1 (0.9%) | 8 (1.4%) | |
| 3 | 3 (0.4%) | 0 | 3 (0.4%) | |
| 4 | 5 (0.7%) | 0 | 5 (0.7%) | |
| 5 | 1 (0.1%) | 0 | 1 (0.1%) | |
| 6 | 1 (0.1%) | 0 | 1 (0.1%) | |
| Trigeminal nerve symptoms | 694/705 | .056 | ||
| None | 659 (95%) | 98 (14.1%) | 561 (80.9%) | |
| Occasionally | 18 (2.6%) | 7 (1%) | 11 (1.6%) | |
| Hypesthesia | 10 (1.4%) | 2 (0.3%) | 8 (1.1%) | |
| Pain | 7 (1%) | 0 | 7 (1%) | |
| Vertigo | 694/705 | .305 | ||
| None | 349 (50.3%) | 47 (6.8%) | 302 (43.5%) | |
| Occasionally | 250 (36%) | 40 (5.8%) | 210 (30.2%) | |
| Permanent | 79 (11.4%) | 16 (2.3%) | 63 (9.1%) | |
| Disabling | 16 (2.3%) | 4 (0.6%) | 12 (1.7%) | |
P-values indicate significant differences. Values are presented as the number of patients (%) and mean ± SD, unless indicated otherwise. Significant P-values (<.05) are highlighted in italics.
The incidence of new neurological symptoms following SRS was significantly higher in the TP-group (20.6% vs 8.4%; ꭓ2 = 14.731, P < .001). Specifically, patients with TP were more likely to develop HFS (15.9% vs 4%; ꭓ2 = 23.364, P < .001). In addition, hydrocephalus was significantly more common in the TP group (1.9% vs 0%; ꭓ2 = 11.209, P < .02) and correlated with tumor size (ꭓ2 = 20.088, P < .001), occurring in 2 cases of T4a tumors that required emergency surgery. The data regarding the onset of new symptoms after SRS are presented in Table 3.
Table 3.
Onset of New Symptoms After SRS
| All N = 705 |
Tumor progression N = 107 |
Tumor control N = 598 |
P-value | |
|---|---|---|---|---|
| Onset of new symptoms after SRS | 72 (10.2%) | 22 (20.6%) | 50 (8.4%) | <.001 |
| Facial nerve affection | 54 (7.7%) | 19 (17.8%) | 35 (5.9%) | <.001 |
| Facial palsy | 11 (1.6%) | 0 | 11 (1.8%) | .157 |
| Hemifacial spasm | 41 (5.8%) | 17 (15.9%) | 24 (4.0%) | <.001 |
| Hydrocephalus | 2 (0.3%) | 2 (1.9%) | 0 | .023 |
| Brainstem edema | 1 (0.1%) | 1 (0.9%) | 1(0.2%) | .169 |
P-values indicate significant differences. Values are presented as the number of patients (%) and mean ± SD, unless indicated otherwise. Significant P-values (<.05) are highlighted in italics.
Tumor Control
The mean follow-up time was 93.6 months in the TC group. TP was diagnosed after a mean period of 47.3 ± 25.1 months (range 9–157 months). At the final follow-up, a significant difference in relative TVs (RTV) was identified between the 2 groups (TC = 0.56 ± 0.8, TP = 2.03 ± 1.9, U = 53760.5, P < .001). A significant difference of the RTV was already found at 12 months (TC = 1.0 ± 0.6, TP = 1.4 ± 1.4, U = 2938.0, P = .002) and even more pronounced after 24 months (TC = 0.71 ± 0.5, TP = 1.5 ± 0.7, U = 5286.5, P < .001, Figure 3A). Failure of reduction of initial TV (RTV > 1) after 24 months was associated with relapse, with a sensitivity of 0.76 and a specificity of 0.85 in the ROC analysis (Figure 3B).
Figure 3.
(A) RTV over time in the TC and TP groups, (B) ROC curve of RTV at 24 months follow-up. Failure of reduction of initial tumor volume (RTV > 1) was associated with TP with a sensitivity of 0.76 and a specificity of 0.85.
Radiation Parameters
The mean marginal dose in the TP group was 12.89 Gy ± 0.32 (median 13.0 Gy) and in the control group 12.96 Gy ± 0.19 (median 13.0 Gy), showing a statistically significant difference (P = .033). Regarding the isodose, our results also showed a difference between the 2 groups with a mean value of 63.41% ± 3.96% in the TP group and 64.30% ± 3.18% in the TC group (P = .035). However, it is worth noting that the difference in the mean marginal dose is less than 0.1 Gy, which is within the 3% dosimetric accuracy and therefore not relevant in clinical practice. The same can be said for the isodose, where a 0.9% isodose difference is too low to have any implications in clinical practice. Other radiation parameters, such as the TaV, PIV, Dmax, D90, and absorbed dose, did not differ significantly.
Our results indicate that the investigated radiation parameters are not predictive of recurrence. In particular, the typical variation in marginal dose between 12 and 13 Gy does not appear to be associated with an increased TP rate.
Predictive Factors of TF
Tumor and patient-specific variables with statistically significant results in univariable analyses were entered into a binary logistic regression model with forward selection. To identify possible predictors of TF before SRS, gender, OHATA class, Tumor size (T1 vs. T2–T4), and tumor configuration (cystic vs. solid) were included. The overall model was significant, showing a good fit of data and correctly classifying 85.7% of cases (ꭓ2 = 15.291, P = .004). The area under the curve was 0.632, demonstrating a sensitivity of 56.3% and a specificity of 73.8%, based on a cutoff value of 0.175. Female gender (OR = 1.72, CI [1.04–2.8]. P = .011) and low OHATA class (OR = 1.44, CI [1.09–1.91]. P = .034) were identified as significant independent predictors of TF.
Discussion
The main purpose of this study was identifying factors predicting TF after Gamma Knife SRS in sporadic VS patients to aid clinical decision-making. SRS has been proven to be a highly effective treatment for VS.3,4,12 However, the course of treatment after SRS is not uniform. Although the majority of VS shrink after a transient increase in tumor size (PP), a small percentage continue to grow. Approximately 9%–12% of VS patients develop TP over time.3,4,6 The present series also confirms these findings, with 15.2% of patients (n = 107/705) experiencing TF. Salvage surgery in these cases can be more challenging due to tumor fibrosis and adhesions to the brainstem, vessels, and cranial nerves.13–15 Subsequently, it might be more difficult to achieve complete tumor resection in comparison to previously untreated VS. Therefore, detection of TF predisposing factors would be of critical importance. In our cohort, 28 patients (26.2%) underwent surgical tumor resection, while 78 (72.9%) have been re-irradiated. Further studies are required to evaluate treatment response after a second SRS.
Patients-related Factors
Our data revealed a statistically significantly higher risk of radioresistance in female patients (P < .05), while patients’ age did not show any influence on TF. VS microenvironment can certainly have an impact through biochemical signaling that may promote angiogenesis, tumor remodeling, or tumor growth.16 Estrogen receptors have been detected in many CNS tumors, including VS. In the past, VS were thought to predominantly affect women, with hormonal influences during menstruation and pregnancy as potential reasons for this gender difference.17,18 Therefore, some authors have studied the expression of estrogen receptors and progesterone receptors in sporadic VS19,20, revealing an upregulation of these receptors and suggesting them as possible therapeutic targets. Nevertheless, additional immunohistochemical studies are necessary since the role of hormones in the pathogenesis and development of VS remains incompletely understood.17,18,20 Based on our findings in a large patient cohort, female gender significantly increases the risk of TF after SRS. This should be taken into account during pretreatment counseling along with other important criteria, such as tumor size and configuration, age, clinical symptoms, comorbidities, and patient preferences, according to a more individualized therapeutic approach. In a recent study focused on the impact of gender on symptoms and quality of life in a cohort of 250 patients, female patients reported a greater postoperative improvement in general health after surgical resection.21 Despite the aim of this research not being to make a comparison between different treatment options (SRS vs surgery), it confirms the importance of gender differences that should be considered when providing pretreatment counseling to VS patients.
Clinical Status and Outcome
The presence of clinical symptoms at diagnosis, such as hearing impairment, dizziness, tinnitus, and facial nerve affection, does not appear to be a useful predictor of TP after SRS. However, a new onset or worsening of these symptoms after irradiation seems to be related to a higher rate of TP (P < .001). In particular, the onset of HFS after SRS seems to be associated with TP (P < .001). Indeed, the incidence of HFS after SRS in the TP group was significantly higher than in the remission group (15.9% and 4%, respectively).
The incidence of HFS secondary to radiosurgery increases significantly in the case of TF. Furthermore, a significant correlation was observed between OHATA Class and HFS, resulting in a higher incidence of HFS in the case of a prominent intrameatal component with lateral growth (Class A > B > C > D). This may be related to a higher intrameatal compression of the FN in OHATA Class A in comparison to the other classes, with a more medial tumor growth. Moreover, low OHATA Class was identified as a significant independent predictor of TP after SRS. This may be related to the necessity for a more cautious radiosurgical planning of intrameatal tumor components, due to their proximity to inner ear structures. Consequently, it is of critical importance to consider the OHATA class in pretreatment counseling with the objective of identifying patients with a higher risk to develop TP and HFS after SRS (OHATA Class A) and therefore more likely to require a salvage surgery to improve FN function.
Tumor Control
TV and configuration.—
The incidence of TF after SRS of 15.2% is comparable to the data in the current literature.22–24 According to our data, TV and tumor configuration did not affect the risk of TF. Regarding TV, our results showed that the risk of TF is not influenced by the volume of the tumor itself. Frischer et al. also described a similar rate of TC among all Koos grades in their cohort of n = 426 patients at long-term follow-up.4 From these data, it appears that there is no difference in TF between smaller (T1 and T2) and larger (T3 and T4) tumors. Rather, it is more likely that the characteristics of the tumor tissue, including biomolecular factors, play a more prominent role. In our cohort, hydrocephalus occurred only in 2 large (T4) tumors after SRS (n = 2/705 cases). Since PP may occur and could lead to brainstem compression or hydrocephalus in large VS,25 surgical resection should be preferred in these cases.3,26
Regarding tumor configuration, our series did not show an increased risk of TP in the case of cystic tumors. However, cystic VS appear to recur significantly earlier than solid VS. The use of SRS as a treatment modality for cystic VS remains a topic of ongoing debate. A meta-analysis including a total of 246 patients suggested that SRS may be a viable treatment option even for cystic VS, achieving effective TC rates.27 However, outcomes of SRS among different studies are difficult to compare and may vary considerably due to the lack of consensus on the definition of “cystic tumors,” which often includes microcystic tumors, as in our series. These may respond better to SRS since the solid component is usually preponderant.
It should be noticed that in our cohort, of the 9 cystic tumors in the TP group, 6 cases required a salvage surgery procedure. Therefore, since cyst enlargement after SRS with consequent brainstem compression or onset of new neurological deficits necessitating surgical intervention may occur, we suggest a cautious approach when considering SRS to treat cystic tumors.
Furthermore, cystic VS appear to recur significantly earlier than solid VS (P < .05), while TV did not influence the time of recurrence. Increased vascular permeability and microhemorrhage may be potential explanations for the accelerated expansion of cystic VS following SRS.28,29
Pseudoprogression.—
In 23%–44% of patients, a transitory tumor growth which spontaneously resolves can be observed after SRS.5 This phenomenon is usually defined as “pseudoprogression” and it is likely to be caused by acute inflammation as a reaction of the tumor cells to radiation, resulting in a temporary enlargement of the tumoral mass. PP typically occurs between 6 months and 1 year after SRS, although some authors have reported a “late” PP, happening up to 3 or 4 years after SRS and then resolving very slowly,7,30 suggesting a more cautious approach in considering salvage surgery to avoid premature treatment.
Despite the existence of some cases in literature in which a new tumor growth was documented many years after SRS and resolved spontaneously, our data, where the mean time of re-treatment was approximately 4 years (47.3 months), demonstrate a clear trend in the absolute TV and RTV. In particular, absolute and relative TV increase significantly in patients with TP as early as 12 months and even more so 24 months after treatment, when compared with the remission group. While these are retrospective data, it can be assumed that in most cases, in patients with a continuous increase in TV, TP should be highly suspected at 12 months and even more at 24 months after SRS. In consideration of the fact that PP is, nevertheless, an increase in TV, and that this situation may extend over many years, it can significantly affect the patient’s quality of life (QoL). Further studies investigating the impact of this condition on the QoL of irradiated patients may be beneficial in determining the best therapeutic approach.
Limitations of the Study
Although we are presenting data in a large study cohort of n = 705 patients, all treated at the same Gamma-Knife center, data collection was performed retrospectively. Moreover, we could not differentiate in our cohort growing from non-growing VS prior to treatment, and since this aspect could be very important in determining the risk of TF, it should be investigated in further studies. Thirdly, the role of the biomolecular aspect in affecting TP was not included in the study design.
Conclusion
Our data indicate that female gender is a risk factor for TF and should be considered in pretreatment counseling, as well as the OHATA class. Furthermore, the latter seems to be related to a higher risk of HFS. Furthermore, the onset of new neurological deficits after SRS treatment, particularly HFS, is significantly more frequent in patients with TP, which could be helpful in suspecting TF. According to the trend of the TV, in case of continuous tumor growth, TP can be strongly suspected as early as 12 months and even more at 24 months after SRS.
Contributor Information
Ludovica Fabbrocini, Department of Neurosurgery, Alfried Krupp Hospital, Essen, Germany.
Helene Hurth, Department of Neurosurgery, Eberhard Karls University, Tübingen, Germany.
Albertus T C J van Eck, Radiosurgery, Gamma Knife Center, Krefeld, Germany.
Marcos Tatagiba, Department of Neurosurgery, Eberhard Karls University, Tübingen, Germany.
Gerhard Horstmann, Radiosurgery, Gamma Knife Center, Krefeld, Germany.
Florian Heinrich Ebner, Department of Neurosurgery, Alfried Krupp Hospital, Essen, Germany.
Funding
The authors declare that this research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Authorship statement
Conception and design: F.H.E. and G.H.; acquisition of data: G.H., L.F., and A.T.C.J.vE.; analysis and interpretation of data: F.H.E., L.F., H.H., and G.H.; statistical analysis: H.H.; drafting the article: L.F. and H.H.; critically revising the article: F.H.E., G.H., M.T., and A.T.C.J.vE..
Conflict of interest statement
The authors certify that there are no personal, financial, or institutional disclosures to be made and there is no conflict of interest.
Data availability
All data and materials are available and can be provided upon reasonable request.
Ethic statement
This retrospective study was conducted in compliance with ethical standards, and it utilized previously collected, anonymized data. Since this study did not involve any interventions or direct contact with patients, and because all data were retrospective and anonymized, informed consent was not required, according to the local ethics committee.
References
- 1. Stangerup SE, Caye-Thomasen P, Tos M, Thomsen J.. The natural history of vestibular schwannoma. Otol Neurotol. 2006;27(4):547–552. [DOI] [PubMed] [Google Scholar]
- 2. Carlson ML, Tveiten OV, Driscoll CL, et al. Long-term quality of life in patients with vestibular schwannoma: An international multicenter cross-sectional study comparing microsurgery, stereotactic radiosurgery, observation, and nontumor controls. J Neurosurg. 2015;122(4):833–842. [DOI] [PubMed] [Google Scholar]
- 3. Watanabe S, Yamamoto M, Kawabe T, et al. Stereotactic radiosurgery for vestibular schwannomas: Average 10-year follow-up results focusing on long-term hearing preservation. J Neurosurg. 2016;125(Suppl_1):64–72. [DOI] [PubMed] [Google Scholar]
- 4. Frischer JM, Gruber E, Schöffmann V, et al. Long-term outcome after Gamma Knife radiosurgery for acoustic neuroma of all Koos grades: A single-center study. J Neurosurg. 2018;130(2):388–397. [DOI] [PubMed] [Google Scholar]
- 5. Thielhelm TP, Goncalves S, Welford SM, et al. Understanding the radiobiology of vestibular schwannomas to overcome radiation resistance. Cancers. 2021;13(18):4575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gugel I, Ebner FH, Grimm F, et al. Contribution of mTOR and PTEN to radioresistance in sporadic and NF2-associated vestibular schwannomas: A microarray and pathway analysis. Cancers. 2020;12(1):177. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Rueß D, Pöhlmann L, Hellerbach A, et al. Acoustic neuroma treated with stereotactic radiosurgery: Follow-up of 335 patients. World Neurosurg. 2018;116(Complete):e194–e202. [DOI] [PubMed] [Google Scholar]
- 8. Samii M, Gerganov V, Samii A.. Improved preservation of hearing and facial nerve function in vestibular schwannoma surgery via the retrosigmoid approach in a series of 200 patients. J Neurosurg. 2006;105(4):527–535. [DOI] [PubMed] [Google Scholar]
- 9. Ohata K, Tsuyuguchi N, Morino M, et al. A hypothesis of epiarachnoidal growth of vestibular schwannoma at the cerebello-pontine angle: Surgical importance. J Postgrad Med. 2002;48(4):253–8; discussion 258. [PubMed] [Google Scholar]
- 10. Yen TL, Driscoll CLW, Lalwani AK.. Significance of house-brackmann facial nerve grading global score in the setting of differential facial nerve function. Otol Neurotol. 2003;24(1):118–122. [DOI] [PubMed] [Google Scholar]
- 11. Mohr G, Sade B, Dufour JJ, Rappaport JM.. Preservation of hearing in patients undergoing microsurgery for vestibular schwannoma: Degree of meatal filling. J Neurosurg. 2005;102(1):1–5. [DOI] [PubMed] [Google Scholar]
- 12. Boari N, Bailo M, Gagliardi F, et al. Gamma Knife radiosurgery for vestibular schwannoma: clinical results at long-term follow-up in a series of 379 patients: Clinical article. J Neurosurg. 2014;121(Suppl_2):123–142. [DOI] [PubMed] [Google Scholar]
- 13. Wise SC, Carlson ML, Tveiten OV, et al. Surgical salvage of recurrent vestibular schwannoma following prior stereotactic radiosurgery. The Laryngoscope. 2016;126(11):2580–2586. [DOI] [PubMed] [Google Scholar]
- 14. Marinelli JP, Herberg HA, Moore LS, et al. Salvage microsurgery following failed primary radiosurgery in sporadic vestibular schwannoma. JAMA Otolaryngol Neck Surg. 2024;150(4):287-294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Lee HJ, Kim MJ, Koh SH, Chang WS, Moon IS.. Comparing outcomes following salvage microsurgery in vestibular schwannoma patients failing gamma-knife radiosurgery or microsurgery. Otol Neurotol. 2017;38(9):1339–1344. [DOI] [PubMed] [Google Scholar]
- 16. Cazzador D, Astolfi L, Daloiso A, et al. Tumor microenvironment in sporadic vestibular schwannomA: A systematic, narrative review. Int J Mol Sci . 2023;24(7):6522. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Jaiswal S, Agrawal V, Jaiswal AK, Pandey R, Mahapatra AK.. Expression of estrogen and progesterone receptors in vestibular schwannomas and their clinical significance. J Negat Results Biomed. 2009;8(1):9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Dalgorf DM, Rowsell C, Bilbao JM, Chen JM.. Immunohistochemical investigation of hormone receptors and vascular endothelial growth factor concentration in vestibular schwannoma. Skull Base. 2008;18(6):377–384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Patel AK, Alexander TH, Andalibi A, Ryan AF, Doherty JK.. Vestibular schwannoma quantitative polymerase chain reaction expression of estrogen and progesterone receptors. The Laryngoscope. 2008;118(8):1458–1463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Brown CM, Ahmad ZK, Ryan AF, Doherty JK.. Estrogen receptor expression in sporadic vestibular schwannomas. Otol Neurotol. 2011;32(1):158–162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Machetanz K, Wang SS, Oberle L, Tatagiba M, Naros G.. Sex differences in vestibular schwannoma. Cancers. 2023;15(17):4365. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Johnson S, Kano H, Faramand A, et al. Long term results of primary radiosurgery for vestibular schwannomas. J Neurooncol. 2019;145(2):247–255. [DOI] [PubMed] [Google Scholar]
- 23. Smith DR, Saadatmand HJ, Wu CC, et al. Treatment outcomes and dose rate effects following gamma knife stereotactic radiosurgery for vestibular schwannomas. Neurosurgery. 2019;85(6):E1084–E1094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Kaylie DM, Horgan MJ, Delashaw JB, McMenomey SO.. A meta‐analysis comparing outcomes of microsurgery and gamma knife radiosurgery. The Laryngoscope. 2000;110(11):1850–1856. [DOI] [PubMed] [Google Scholar]
- 25. Huang CW, Tu HT, Chuang CY, et al. Gamma Knife radiosurgery for large vestibular schwannomas greater than 3 cm in diameter. J Neurosurg. 2018;128(5):1380–1387. [DOI] [PubMed] [Google Scholar]
- 26. Tatagiba M, Wang SS, Rizk A, et al. A comparative study of microsurgery and gamma knife radiosurgery in vestibular schwannoma evaluating tumor control and functional outcome. Neuro-Oncol Adv. 2023;5(1):vdad146. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Ding K, Ng E, Romiyo P, et al. Meta-analysis of tumor control rates in patients undergoing stereotactic radiosurgery for cystic vestibular schwannomas. Clin Neurol Neurosurg. 2020;188:105571. [DOI] [PubMed] [Google Scholar]
- 28. Frisch CD, Jacob JT, Carlson ML, et al. Stereotactic radiosurgery for cystic vestibular schwannomas. Neurosurgery. 2017;80(1):112–118. [DOI] [PubMed] [Google Scholar]
- 29. Park CK, Kim DC, Park SH, et al. Microhemorrhage, a possible mechanism for cyst formation in vestibular schwannomas. J Neurosurg. 2006;105(4):576–580. [DOI] [PubMed] [Google Scholar]
- 30. Breshears JD, Chang J, Molinaro AM, et al. Temporal dynamics of pseudoprogression after gamma knife radiosurgery for vestibular schwannomas—A retrospective volumetric study. Neurosurgery. 2019;84(1):123–131. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data and materials are available and can be provided upon reasonable request.