Background: Vestibular schwannomas (VSs) are benign nerve sheath tumors that arise from the vestibulocochlear
nerve within the internal auditory canal (IAC) and extend into the cerebellopontine
angle (CPA). It is the most common tumor of the CPA with an annual incidence of 17.4/1
million. They typically demonstrate slow growth over time and as such, observation
is a reasonable approach to management. A portion of these tumor remains static and
approximately 5 to 10% of these tumors will demonstrate spontaneous regression while
under observation, including those associated with neurofibromatosis type 2. The standard
treatment for symptomatic or growing lesions is surgical resection followed by radiation
for residual tumor or reoccurrence; however, management recommendations encourage
tailoring care to each patient and tumor individually.
Several previous case series have attempted to identify predictive factors for tumor
growth and regression, but few have reached significance or demonstrated reproducible
findings. If factors were identified that could predict tumor growth one could intervene
at an earlier stage of the disease. One meta-analysis conducted on growing tumors
identified that size at diagnosis was the only predictive factor that reached significance.
In a similar vein, defining factors that reliably predict spontaneous regression could
prevent unnecessary intervention. As per our review of the literature, no patient
characteristics have yet predicted spontaneous regression to date. Imaging characteristics
including a festooned aspect of the tumor and the presence of cerebrospinal fluid
in the IAC have been identified as predictive for tumor regression in small case series
(N = 13–14).
Methods: Using a clinical database of VS treated by one team at our institution, we identified
40 patients who have demonstrated significant spontaneous regression or complete resolution
of their VS. All patients received a survey by mail and telephone. For all patients
who consented to participate, radiographic and clinical data was collected from patient
charts in addition to survey responses. Medical comorbidities and medications provided
through patient questionnaire were corroborated with patient charts. Tumor volume
was approximated using the formula V = 4/3 × π × length/2 × width/2 × height/2 and
nominal logistic regression was completed using JMPv17 with 50% tumor reduction as
the reference value.
Results: Ten patients were included in the final descriptive summary of this patient population.
Tumors were generally small, left sided (8/10), with a fungated shape (6/10) and at
least partial preservation of CSF in the IAC (8/10). Significant decrease in tumor
volume was an inclusion criterion; however, no patient factors or radio graphic factors
appeared to predict the degree of tumor regression.
Conclusion: In conclusion, this is the first study to consider patient lifestyle factors obtained
through patient survey in addition to clinical and radiographic findings to describe
spontaneous regression of VS. Although there was perhaps a higher-than-average rate
of herpes/varicella in this population, no clinical factors were found to be predictive
on analysis. Additionally, no radiographic factors appear to be protective including
those previously demonstrated to be protective, such as CSF preservation in the IAC
and IAC extension.
