Sex-specific difference in treatment success/failure after vestibular schwannoma treatment

Abstract

Background

Sex-related differences in patients with sporadic, unilateral vestibular schwannoma (VS) are poorly investigated so far, and it remains unclear whether biological sex affects treatment response to stereotactic radiosurgery (SRS) or microsurgical resection (SURGERY). This study elucidates sex-related differences in treatment outcome of VS.

Methods

This is a retrospective two-center cohort study. All consecutive patients treated for their VS between 2005 and 2012 were included. Previously treated VS and patients with neurofibromatosis were excluded. Clinical status and treatment-related complications were analyzed from both centers’ prospective treatment registries. Recurrence/progression-free-survival was assessed radiographically by contrast-enhanced magnetic resonance imaging.

Results

Within the entire patient cohort of N = 1,118, the majority of VS patients (56%) was female. Sixty-two percent of patients were treated by SRS. Females with very small tumors (KOOS I) were significantly less likely to be assigned to SURGERY than males (P = .009). Mean follow-up time was 6 ± 4.3 years. In SURGERY, the rate of subtotal resection was significantly higher in women (7%) compared to men (2%) (P = .041). However, there was no difference in long-term tumor control after SURGERY between both sexes (P = .729). In SRS however, the incidence of recurrence was significantly higher in women (14%) compared to men (8%) (P = .004), which was also reflected in the Kaplan–Meier analysis (P = .031).

Conclusions

Female sex was a negative prognostic factor for treatment success (long-term tumor control) if treated with SRS—there was no sex-related differences in long-term tumor control after SURGERY. Additional research is needed to elucidate sex-related differences in tumor biology affecting the response to VS treatment.

Vestibular schwannoma (VS) are benign intracranial schwannomas arising from the eighth cranial nerve.^1–3 Possible treatment options for VS include stereotactic radiosurgery (SRS), microsurgical resection (SURGERY), “Wait-and-Scan-Strategy” or combined approaches.⁴ Many studies have compared treatment effectiveness between the different modalities, however, surrogate markers to predict treatment success/failure beyond tumor size and cystic morphology are currently lacking.^3,5–8 Nevertheless, investigating predictive parameters to anticipate the responder status to SRS is crucial to successfully provide personalized recommendations on VS management in the future.

Sex differences in both incidence and outcome have been discussed in the field of neuro-oncology and neurosurgery in the past, eg, longer survival in female glioblastoma patients or higher incidence of meningioma and pituitary tumors in females.^9–16 The prevalence of VS has been described to be more frequent in females compared to males, and a population-based case–control study on VS risk factors has revealed an elevated risk for larger VS in pregnant females, and a higher risk for VS in parous compared with nulliparous females.^4,17–21 These findings suggest that there might be underlying pathophysiological differences in VS affected by biological sex.^4,17–21 This study aims to elucidate sex-related differences in VS presentation, management, and treatment response (treatment failure and success) in a comparative study design both after SRS and SURGERY.

Methods

Study Design and Patient Cohort

This is a retrospective two-center cohort study. Patients were identified by a prospectively kept registries of two tertiary and specialized centers involved in the treatment of VS. Data were then retrospectively collected for patients treated between 2005 and 2012 to enable long-term follow-up. Previously treated VS and VS associated with Neurofibromatosis were systematically excluded. The local ethics committee approved this analysis and was in accordance with the ethical standards laid down in the Declaration of Helsinki.

Data Collection

Tumor size was classified by KOOS Classification.²² Facial function was reported by House and Brackmann (H&B) scale (with H&B 1-2 considered as good outcome) and Gardner-Robertson (G&R) scale (G&R 1-2 considered as good outcome). Peri-interventional complications were classified based on the Clavien-Dindo-Classification (CDC).²³ Recurrence-free survival was assessed radiographically by contrast-enhanced magnetic resonange imaging (MRI).^24,25 The criteria for tumor recurrence/progression was progredient growth in Gadolinium contrast-enhanced MRI. In the SRS treatment arm, only growth past 2 years after the treatment was graded as recurrence to exclude known phenomenon of pseudoprogression.²⁶ Treatment success was defined as progression/recurrent-free survival.

In case of SURGERY, extent of resection (EOR) was classified by first post-operative MRI (3 months postoperative): residual contrast-enhancing tumor was defined as subtotal resection (STR), whereas gross total resection (GTR) was defined as lack of contrast-enhancement in Gadolinium-enhanced MRI.

Treatment Modalities

Patients treated by SURGERY were all operated by retrosigmoid approach using intraoperative electrophysiological monitoring. Patients were either operated in semi-sitting or supine position.²⁷ All VS patients in the SRS cohort received Gamma-Knife-Radiosurgery (GKR—Elekta AB, Stockholm, Sweden) with a prescription dose of 13 Gy to the 65% isodose line.

Statistical Analysis

Statistical analysis was performed in R Studio (Version 1.2) using descriptive statistics. To compare nonnumeric parameters of both groups, the Fisher’s exact test was applied. For numeric parameters, Welch’s 2 sample t-test was used. Recurrence-free survival was estimated using the Kaplan–Meier method and compared between cases and controls using a log-rank test. The length of follow-up for recurrence-free survival was calculated from the date of surgical or radiosurgical intervention to the date of either recurrence or the last clinical visit. Significance was defined as the probability of a 2-sided type 1 error being < 5% (P < .05). Data are presented as mean ± standard deviation (SD) if not indicated otherwise.

Results

Patient Demographic and Provided Care

In this patient cohort of N = 1,118, most VS patients were female with 56%. Tumor size distribution was similar in both sex groups and are shown in Figure 1A. At time of treatment, females were significantly older in age compared to males (P = .003). Overall, the majority was treated with SRS (62%) there was no difference in both sex groups (P = .072) (Table 1). When comparing provided care modality in either sex within same sized tumors, there was a significant lower rate of choice for SURGERY for females with 14% in KOOS I compared to males with 32% (P = .009) (Figure 1B).

Females treated with SURGERY were significantly younger (mean 49 ± 12.32 years) compared to SRS-treated females (mean 59 ± 13.04 years) (P= < .001; CI:8.24–12.44). This phenomenon was also present in the male cohort (SURGERY: 47 ± 12.36 years; SRS: 58 ± 12.76 years (P = .001; 95CI:8.22-12-76). The overall incidence of cystic morphology and shunt-dependency are shown in Table 1 and are indifferent in either sex group. Treatment-related side-effects and complication classified in CDC are shown in Table 1 and are indifferent in both sex groups. When treated with SURGERY, 70% of all female patients were operated in semi-sitting position, similar to male patients (69%) (P = .814). The rate of STR was significantly higher in females compared to males (7% and 2%, P = .041). The treatment parameters were indifferent in either sex group after SRS: Conformity (Paddick) index (P = .246), therapeutical dose (P = .210), and maximal dose (P = .148). However, the tumor volume was slightly higher in males compared to females (1.70 ± 2.38 ccm and 1.35 ± 1.64 ccm, P = .027; 95CI:0.04-0.64).

Functional Status and Postoperative Outcome

The pre-operative incidence of facial palsy, hearing deterioration, tinnitus, and trigeminal affection at treatment timepoint was indifferent between both sexes. However, females significantly more often suffered from symptomatic vertigo (Table 2). The rate of good facial function preservation (HB 1-2) after SURGERY was analogous in both sexes (88% and 91%, P = .426). Hearing preservation (G&R1-2) was also not significantly impacted by patient’s sex (33% and 41%, P = .197). Postinterventional good facial preservation after SRS was independent of sex (99% and 99%, P = 1). Between both treatment modalities, hearing preservation after SRS was significantly higher compared to SURGERY (54% and 37%, P < .001), but there were no sex-related differences after SURGERY (51% and 59%, P = .145). There was no difference in treatment-related adverse effects between both sexes (Table 1).

Treatment Response Regarding Tumor Control

Mean follow-up time was 6 ± 4.26 years in the overall cohort. Overall, the incidence of recurrence was significantly higher in females compared to males (P = .027). However, this difference was due to the SRS-cohort with a treatment failure rate of 14% in females compared to 8% in males (P = .004) (Table 1). As follow-up time was comparable in both sex groups of SRS-treated VS (6 ± 4.31 years and 6 ± 4.12 years, P = .406), this result was not biased by difference in follow-up time. From all progressive/recurrent VS after SRS, 14% received surgical resection and 86% received second SRS.

Kaplan–Meier analysis considering time and tumor control in the general cohort revealed no significant difference in overall progression-free survival (Figure 2A). In contrast, treatment success rate depended on tumor size (classified in KOOS Classification) in both SRS and SURGERY (Figure 2B).

In patients treated with SRS, however, females faired significantly worse considering tumor control compared to males (P = .031) (Figure 3A). In patients treated with SURGERY, there was no sex-related difference in treatment success rate nor progression-free-survival (Table 1 and Figure 3B).

Mean time to recurrence was 6 ± 3.45 years overall but significantly shorter after SRS with 5 ± 3.25 years compared to SURGERY with 7 ± 3.43 years (P = .002; 95CI:0.87–3.57). However, within both treatment arms, there were no sex-related differences in mean time to recurrence with 5 ± 3.35 years in SRS-treated females, 5 ± 3.06 years in SRS-treated males (P = .664) and 8.±3.33 years in SURGERY-treated females and 7 ± 3.48 years in SURGERY-treated males (P = .218).

Discussion

In our two-center cohort of N = 1,118 patients treated for their VS, we have found that sex-related differences in provided care, clinical presentation, and incidence for treatment success/failure exist. When treated with SURGERY, females were more likely to receive STR instead of GTR. In KOOS I tumors, females were likely to receive SRS instead of SURGERY. What is more, female sex was associated with a lower probability for progression-free-survival and higher rate for treatment failure compared to males, when treated with SRS, even though except for more advanced age, there were no significant difference in tumor and patient characteristics. To the best of our knowledge, this is the first analysis to uncover this sex-related difference in treatment response after SRS in VS.

Sex Differences in Provided Care

Both the lower rate of GTR and the lower probability to receive SURGERY in very small VS in female patients point out provided care biases toward less aggressive treatment options in females. It remains to be speculated if the reason for this significant discrepancy is a form of “gender-bias” or “provider-bias”: Previous studies have reported that not only does patients’ sex but also physicians’ sex play a role in the decision-making process in health care, showing a tendency for less aggressive treatments for female patients.^28–31

Moreover, as VS usually presents with benign behavior and functional preservation is of high priority, and we cannot exclude that intention to treat was different in both genders groups considering the higher rate of STR after VS SURGERY. It has been shown that facial nerve deterioration remains one of the most important parameters predicting patient’s quality of life.^32,33 It is possible that female patients or care providers prioritized facial function preservation in females higher (may it be explicit patient’s wish or subconsciously) resulting in a less aggressive surgical course with higher risk for residual tumor. However, the retrospective nature of this study design does not allow a definite answer to this observed phenomenon and the intention-to-treat. Moreover, other gender groups may have to be investigated in the future.

Female Sex as a Negative Predictive Factor for Treatment Success

According to this analysis, the risk for treatment failure after SRS is significantly higher in biological female sex despite same treatment parameters and even significantly lower tumor volume in SRS-treated females (compared to males). The follow-up time was comparable in both sex groups, therefore, we can rule out that a bias based on difference in length of follow-up or loss of follow-up resulted in this phenomenon. Interestingly, SURGERY-treated females did not fare worse compared to males considering tumor control after SURGERY, even though the risk of residual tumor was higher—suggesting that this negative predictive effect is SRS-specific. In studies focusing on natural history, tumor growth rate was described to be indifferent in either sex.³⁴ Therefore, even though the pathophysiology is unclear, we hypothesize based on the current literature that the difference in treatment response after SRS indeed represent a difference in response to SRS based on biological female sex. In radiotherapy and radiosurgery, biological sex has been found to be a prognostic factor for survival and radiosensitivity.^35–38 Endocrinology, inflammatory response, genetic, developmental, and anatomical differences could be accredited for this sex phenomenon.^35,39–42 Previous studies have investigated the differences in hormone receptors in VS, which have reported conflicting results with some studies reporting a negative status of estrogen/progesterone receptors and others reporting histological positivity of estrogen/progesterone receptors or positivity on a mRNA-level.^43–46 Still, the investigation of genetic, molecular, or immunological pathways has to be carried out in the future similar to current neuro-oncological investigations in glioma, to better understand the multilevel sex differences in VS.^47–50 Differences in immune response, mutation clonality, and neuronal signaling pathway have suggested to pursue sex-specific treatment approaches in glioma.^48–50

Also, with these clinical findings in VS, one has to consider to adapt the treatment strategy (ie, SRS treatment plan) or recommendation according to the patient’s sex in VS. The clinical importance of the negative predictive factor of female sex in SRS carries even more weight, when considering both the higher prevalence of VS in women and the higher rate of less-aggressively treated female VS patients (provided care difference). It is important to have in mind that treatment of pretreated VS remains more challenging with a higher morbidity. Onco-functionally safe second or third-line treatment after failed VS treatment still must be defined and found. Therefore, reducing the primary rate of unsuccessful treatment should be the highest aim. Indeed, biological sex has until now not been included in any guidelines for vestibular schwannoma management⁴—however, this data suggests that it should be considered, when discussing SRS as primary treatment choice.

Strength and Limitations of This Study

The strength of this study is its two-center design of two institutions highly specialized in the treatment of VS (by SRS and SURGERY) involving a large number of patients. Mean time of follow-up was longer than 6 years and therefore rather high. However, due to the retrospective nature of our study, our study also bears limitations. First, the association of provided care difference cannot be linked to a causality in this retrospective study design. Second, to rule out social or lifestyle parameters as a cause in provided care, a prospective study design is needed. Furthermore, molecular and histopathological parameters have to be compared between both sexes in the future to better understand the pathophysiology behind our described sex-related difference in treatment response after SRS.

Conclusion

Functional outcome and peri-interventional morbidity were the same in either sex. Sex-related differences in provided care exist with women more likely to be treated with less aggressive therapy. Moreover, female sex was a negative prognostic factor for treatment success (long-term tumor control) if treated with SRS—there was no sex-related difference in long-term tumor control after SURGERY. Additional research is needed to elucidate sex-related difference in tumor biology affecting the response to VS treatment and patient’s sex have to be considered as an important factor, when tailoring treatment strategy in VS.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Conflict of interest statement. The authors report no conflict of interest.

Authorship Statement

Conception and design: S.S.W. and G.N. Acquisition of data: S.S.W.; AvE., and G.H. Analysis and interpretation of data: S.S.W.; G.N. and M.T. Statistical analysis: S.S.W. and G.N. Drafting the article: S.S.W. Critically revising the article: AvE; G.H.; M.R., and M.T. Reviewed submitted version of the manuscript: G.N.

Data Availability

All data and materials are available and can be provided upon reasonable request.

References