Upfront Radiosurgery for Treatment of Symptomatic Obstructive Hydrocephalus due to Brain Tumors

Introduction Hydrocephalus is a build-up of cerebrospinal fluid (CSF) in the brain and is characterized by abnormal dilatation of the cerebral ventricles. Patients can be either asymptomatic, have symptoms related to primary tumors, or have hydrocephalus-related symptoms. Generally, symptomatic patients are candidates for ventriculoperitoneal (VP) shunt placement to reduce acute symptoms. Little evidence exists regarding the resolution of symptomatic hydrocephalus secondary to brain tumors using stereotactic radiosurgery (SRS) alone as a primary treatment option. Methods The present study is a retrospective series of eight patients (six men and two women) diagnosed with obstructive hydrocephalus due to brain tumors treated with radiosurgery between April 2013 and February 2021. The primary endpoint of the present study is to report our institutional experience regarding the control of symptomatic obstructive hydrocephalus due to brain tumors treated with upfront radiosurgery. Results The mean age was 52 years (range, 5-79). The most common presenting symptoms included headache (100%), vision-related symptoms (75%), and ataxia (37.5%). All patients showed symptom improvement after radiosurgery, five (62.5%) patients showed resolution in less than three days and the rest of the patients resolved hydrocephalus in a longer timeframe (more than three days). All patients lowered their Evans index compared to the index documented before radiosurgery, in a range from 0.02 to 0.17. Conclusion Radiosurgery is a non-invasive alternative treatment for primary and secondary brain tumors that debut with obstructive hydrocephalus, tumors expected to have a high alpha/beta ratio might be suitable to attempt radiosurgery to avoid permanently implanted devices such as VP shunts or other invasive procedures such as a third ventriculostomy. The present study demonstrated that in selected cases SRS can lead to hydrocephalus symptom resolution along with a decrease in ventricular size in a relatively short time frame. Little evidence exists regarding the effect of SRS on symptomatic hydrocephalus resolution and further histology-specific studies are required. We acknowledge that this approach requires immediate access to radiosurgery and close clinical follow-up to ensure success.


Introduction
Hydrocephalus is a build-up of cerebrospinal fluid (CSF) in the brain and is characterized by abnormal dilatation of the cerebral ventricles. The CSF accumulation may be due to obstruction in the normal flow or an imbalance in production and absorption [1]. The obstruction most commonly occurs at the foramen of Monro, the aqueduct of Sylvius, the fourth ventricle, and the foramen magnum; most tumors with a significant size or that produce a mass effect by edema can impede circulation at any point of CSF pathways causing obstructive hydrocephalus [2]. Frequent tumors associated with obstructive hydrocephalus include ependymoma, subependymal giant cell astrocytoma, choroid plexus papilloma, craniopharyngioma, pituitary adenoma, hypothalamic glioma, hamartoma, and metastatic tumors. Similarly, posterior fossa tumors are a common cause. Obstructive hydrocephalus also presents in 70% of tumors located in the pineal region [3].
Patients can be either symptomatic, have symptoms related to the primary tumor, or have hydrocephalusrelated symptoms [1]. Generally, symptomatic patients are candidates for VP shunt placement to reduce acute symptoms. Direct removal of the lesions, endoscopic third ventriculostomy (ETV) plus biopsy, and additional shunt procedures are the most common treatment options. The primary disease diagnosis directs the subsequent management and follow-up strategy for these patients after hydrocephalus has been resolved [4]. SRS has proven to substantially reduce tumor volume in short intervals of time in tumors that are expected to have high alpha/beta ratios [5,6]. In symptomatic large brain metastases from lung and breast histology, substantial symptom improvement or resolution can be typically seen in 72 hours in the majority of patients [5], this can also be seen in primary brain tumors with expected high alpha/beta [6]. Thus, the rationale for attempting to cure secondary hydrocephalus by quickly shrinking the tumor causing the CSF obstruction with radiosurgery might be worth exploring, with the aim to spare an invasive procedure such as a permanent implant of a VP shunt. In a previous report from our institution using two-staged radiosurgery for primary large brain tumors, two patients presented with symptomatic obstructive hydrocephalus; giving upfront radiosurgery avoided the need for shunt placement or any other invasive intervention as quick tumor response to radiosurgery was obtained thus resolving the secondary hydrocephalus [6]. We reviewed our case series of patients that presented with symptomatic hydrocephalus secondary to brain tumors suspected of having high alpha/beta ratios that were managed with upfront radiosurgery as a sole treatment alternative.
To our knowledge, this is the first study to report early improvement of symptomatic obstructive hydrocephalus associated with brain tumors using SRS. There is little evidence regarding the resolution of symptomatic hydrocephalus secondary to brain tumors using SRS [6][7][8][9][10]. The primary endpoint of the present study is to report our initial experience regarding the resolution of symptomatic obstructive hydrocephalus due to brain tumors treated with radiosurgery.

Materials And Methods
The present study is a retrospective series of patients diagnosed with symptomatic obstructive hydrocephalus due to brain tumors (hydrocephalus was defined as Evans index >0.30) treated with either single session or adaptive/two-session radiosurgery technique. Between April 2013 and February 2021 we reviewed patient's clinical records that met the following criteria: patients treated with radiosurgery with either a Linear Accelerator (LINAC) (TomoTherapy Incorporated, Madison WI, USA) or a rotating gamma-ray unit (Infini, Masep Medical Company, Shenzhen, China). Both genders were included, with no age restriction for patients presenting with symptomatic obstructive hydrocephalus due to primary or secondary brain tumor documented on magnetic resonance imaging (MRI), and follow-up of at least one month with MRI after radiosurgery. Criteria for exclusion included patients with incomplete medical records, without follow-up MRI or lost to follow-up.
A set of variables were considered among all patients: for the clinical profile, we included gender, age, diagnoses, anatomic area of brain tumor, and presenting symptoms. The radiosurgical plan characteristics comprehend the date of radiosurgical treatment, type of radiosurgical plan, tumor volume, and radiation dose delivered. And for clinical evolution, we measured Evans index prior to and after radiosurgery, response evaluation criteria for solid tumors (RECIST), follow-up MRI, time of symptom resolution with a cut-off point of three days, patients' actual status (alive or deceased), and progression-free survival.
The Evans index was calculated from the ratio of the maximum width of the frontal horns of the lateral ventricles (A) and the maximal internal diameter of the skull (B) at the same level in axial MRI, performed during the radiosurgery simulation process and from follow-up MRI (Evans index = A/B). The RECIST scale ( Table 1) was obtained by measuring the percentage of tumor reduction in brain tumor volume documented during the simulation process and follow-up MRI.  All cases were discussed and approved by a multidisciplinary team of radiation oncologists, medical physicists, and neurosurgeons. All patients and caregivers signed informed written consent. The study protocol was approved by the International Cancer Center Group ethical committee board.

FIGURE 2: Tumor and hydrocephalus evolution for patient six
Top row: prior to SRS.
(A) Axial T1 gadolinium-enhanced MRI the day before treatment showed a large hypothalamic glioma. PTV in this plan (7.3 cc) was drawn within the GTV (8 cc). The prescription dose for PTV was 13 Gy at a 50% isodose curve and GTV received a median and maximum prescription dose of 0.5 and 1 Gy, respectively. (B) Axial T2 MRI on the day of treatment, the image shows dilation of the lateral ventricles, the transpendymal edema and a deviation of the septum pellucidum due to hydrocephalus. EVAN'S index at that moment was 0. 34.
Bottom row: Second treatment session of two-staged radiosurgery (30 days after).

TABLE 2: Patient characteristics
Regarding radiosurgical plan characteristics, two different types of radiosurgical platforms were used: Tomotherapy, which was used for the treatment of metastasis in one patient using a single treatment session, and Infini from which two different radiosurgical plans were used: adaptive or two-stage radiosurgery with a 30-day interval between sessions and single-session radiosurgery depending on each tumor characteristics. Table 3 shows details of radiosurgical doses and tumor volume.

Gamma Knife (Infini)
Adaptive radiosurgery   Table 4 shows a summary of key clinical data for each patient and hydrocephalus response to radiosurgery. All patients had daily follow-up for the first five days and every week thereafter to measure symptomatic response to treatment; all patients showed symptom improvement after radiosurgery, five (62.5%) patients showed resolution in less than three days and the rest of the patients resolved hydrocephalus in a longer timeframe (more than three days). The mean Evans index prior to radiosurgery was 0.35; after radiosurgery mean Evans index was 0.26 (0.09). During this three-day interval, a clinical correlation between symptom improvement and follow-up MRI was analyzed. It is worth mentioning specific cases: patient one was highly disoriented, ataxic, and had decreased strength in his lower limbs; 24 hours after radiosurgery, symptoms had improved significantly, and a follow-up MRI showed improvement in hydrocephalus. Contrarily, patient five took a longer time to recover from presenting symptoms and needed more than three days to completely resolve symptomatic hydrocephalus; the differential of Evans index in her case was only 0.02. A significant difference between these two cases is that patient five underwent ventriculostomy months prior to radiosurgery, contrary to patient one who was only treated with radiosurgery. Patients who had complete or partial tumor response (RECIST 1-2) after radiosurgery showed a reduction of Evans index below 0.31 (75% of the patients). Patients with stable disease (RECIST 3) did not lower their Evans index below 0.31. Despite this, all patients lowered their Evans index compared to the index documented before radiosurgery, in a range from 0.02 to 0.17. At the last follow-up, three patients were deceased, two from primary disease progression, and the other patient developed peritoneal tumor seeding after VP placement. The mean time of progression-free survival among these patients was 41 months (range 1-103).

Discussion
The present study, in contrast to routine neurosurgical management of secondary hydrocephalus, provides evidence of the effectiveness of SRS for the acute management of symptomatic obstructive hydrocephalus in selected cases of brain tumors with an expected high alpha/beta ratio. The linear-quadratic (LQ) model has been validated by experimental and clinical data as an equation that can aid to predict biological response after irradiation. The LQ describes the surviving fraction (SF) of clonogenic or stem cells as a function of radiation dose. The alpha and beta represent the intrinsic radiosensitivity of the cells, thus a higher alpha and beta of a tissue implies that the tissue or tumor is more sensitive to radiation, typically fast-growing tumors such as metastasis, are assigned the highest alpha-beta of ten [11]. Our previous work like those of others has demonstrated that such tumors in the majority of cases will shrink significantly in a relatively brief amount of time (days to weeks) [6]. Thus the rationale for treating secondary hydrocephalus caused by tumors known to have high alpha-beta ratio is based on the expected radiosensitivity and quick response to a high dose of radiation such as those delivered by single fraction radiosurgery with the aim of treating the cause (tumor) and not the consequence (hydrocephalus).
Comparison of the Evans index at different time frames showed a satisfactory response to treatment through the evident reduction of ventricle size by one to two months after SRS, showing no signs of recurrence until the last follow-up; patients who underwent two-staged radiosurgery showed a significant decrease in hydrocephalus by the second radiosurgical treatment (30 days after). Radiation to the tumor-causing obstruction at the third or fourth ventricle among our patients contributed to improving cerebrospinal fluid flow and resolution of hydrocephalus promptly, most tumors reduced their volume significantly in the short term. Our study findings show that SRS is an effective treatment alternative to improve symptomatic obstructive hydrocephalus in the early and acute phases of the disease and this is maintained in the long term in selected patients. The rapid response and early effect of SRS rely on the histology of the tumor and its radiosensitivity (high alpha-beta) [5,6,[12][13][14].
A previous study reported four patients with posterior fossa metastasis causing obstructive hydrocephalus [15]. Rapid rescue radiosurgery in a three-staged session with a prescription dose ranging from 6-8.5 Gy for a 35% isodose line per treatment was performed for a period of seven days intending higher prescription dose to the central tumor [15]. All patients achieved a favorable response of obstructive hydrocephalus and positive tumor regression. Matsunaga et al. [9] reported on the effectiveness and safety of GKSRS for asymptomatic obstructive hydrocephalus associated with posterior fossa metastasis. Radiosurgical plans varied from single up to five fractions and doses ranging from 16-35 Gy depending on each tumor's characteristics. The Evans ratio at GKSRS (median 0.31) improved significantly compared with that at one-three months after GKSRS (median 0.26) and maintained at 6-12 months follow-up. All but one patient avoided surgical procedures for hydrocephalus after GKSRS. Within their findings, primary gastrointestinal cancer, single, two, and three staged radiosurgery, higher age (>65), and male sex correlated with unfavorable outcomes [9].
Among our patients, one had colorectal adenocarcinoma with metastasis to the posterior fossa, aged >65 years. Although he was programmed for adaptive radiosurgery he passed away 10 days after his first session from abdominal sepsis, nevertheless, we managed to document a radiological cure of his hydrocephalus by Day 8, which was consistent with the family report of alleviation of hydrocephalus symptoms in 48 hours after treatment and what they considered a complete resolution of symptoms on the third day. Our institutional experience with two-staged radiosurgery to large primary brain tumors [6] and brain metastasis [5] showed a mean volume reduction of 73% and 66% of lesions between sessions, respectively. As discussed in a previous study, higher intratumoral doses such as those delivered by GKRS to the lesion might correlate with prompt symptom relief including symptomatic obstructive hydrocephalus [5]; although the significance of a lower intratumoral dose variation such as those that are generally achieved with a LINAC under the same prescription dose and its effect on symptom alleviation needs to be proven. Among our patients, the only one who received SRS with LINAC, also diagnosed with brain metastasis, showed a reduction of her Evans index by 13% with a complete resolution of symptoms in less than four days and complete regression of hydrocephalus during follow-up.
On the other hand, Gutierrez et al. [9] described a minimally invasive technique to perform a radiosurgical third ventriculostomy using LINAC in a patient with mild obstructive hydrocephalus secondary to renal cell carcinoma. One week after irradiation, the patient showed improvement in presenting symptoms including diplopia, quadriparesis, and dysphagia. CT scans did not show any changes in the metastatic lesions, even in the peritumoral edema, however, the ventricular size index (VSI) diminished by 4% (from 36% in pre radiosurgery to 32% at a week post-treatment) despite persistent aqueduct obliteration. In the follow-up head CTs, the ventricular index was maintained between 30% and 32% without further need for surgical intervention. Given the effectiveness of the procedure, the authors performed a second radiosurgical ventriculostomy recently [16]. Among our patients, the only one who received SRS with LINAC, also diagnosed with brain metastasis, showed a reduction of her Evans index by 13% with a complete resolution of symptoms in less than four days and complete regression of hydrocephalus during follow-up.
Tumors in the pineal region are usually accompanied by obstructive hydrocephalus. Common presenting symptoms among these patients include headache, gait disturbance, diplopia, and Parinaud syndrome, which is a sign of acute obstructive hydrocephalus. Li et al. [17] reported 147 cases of pineal region tumors treated with GKSRS; 16 patients within their series presented with obstructive hydrocephalus and had a VP shunt was placed to prevent perioperative and postoperative intracranial hypertension. Findings suggest that the safety of GKSRS can be improved by actively dealing with hydrocephalus prior to radiosurgical treatment. Bechri et al. published a case report on a patient with a papillary tumor of the pineal region presenting with obstructive hydrocephalus; the patient underwent ETV, which showed improvement of initial symptoms, the disappearance of nausea and vomiting, and only slight improvement in visual symptoms [18]. The tumor was treated with GK ICON with a dose of 14 Gy at a 50% isodose and the patient was followed up at three, six, and 12 months thereafter. During follow-up, the neurological examination showed progressive improvement until showing complete resolution of symptoms after radiosurgical treatment; tumor reduction and aqueduct compression disappeared after six months [18]. Zhang et al. [19] argued on the existing controversial management regarding when and how to treat hydrocephalus secondary to pineal region tumors. Authors of several clinical studies believe that VP shunt or external ventricular drainage should be performed prior to surgical intervention in order to reduce intracranial pressure and improve the general condition of patients; others believe that preoperative shunting might lead to ventricular collapse and poor tumor exposure, rendering surgical access more difficult and total tumor removal less likely [20][21][22]. Authors in this series suggest no preoperative VP shunting or external ventricular drainage should be performed before tumor resection unless the patient presents with a severe intracranial hypertension crisis; such decisions carry associated risks of infection and malfunction, and the incidence of shunt failure is significant [23,24]. One of our patients was initially diagnosed with pineocytoma, received radiosurgical treatment, and had a good response by improving symptoms. A year after SRS, he underwent a surgical procedure for VP shunt placement due to disease progression. He further developed peritoneal tumor seeding, which was biopsied, revealing a germinoma, ultimately causing his death.
The present study demonstrated the effectiveness of upfront SRS in symptomatic hydrocephalus secondary to brain tumors with expected high alpha-beta ratios. In our series, we presented patients with heterogeneous diagnoses with obstruction in different locations of the cerebrospinal fluid flow, all showing reduction in their Evans index. Although not all patients presented resolution of hydrocephalus (Evans index below 0.31) all of them presented a clinical resolution in a period of days to a week after SRS. Based on our findings, radiosurgery can be considered as an alternative approach instead of a VP shunt or third ventriculostomy in symptomatic hydrocephalus due to brain tumors with an expected rapid response to radiation. Of course, this requires a close follow-up of symptoms evolution, hospitalization if necessary, and immediate access to radiosurgery. Further research is necessary regarding this topic to generate safer treatment protocols for this set of patients.