ITD is a rare complication with an incidence of 4.4% in our study. Due to additional radiotherapy, the prognosis of ITD is better than that of the generally reported LMD [4–8]. Furthermore, young age and good KPS are good prognostic factors for ITD.
The pathogenesis of LMD in patients with GBM remains unclear. Tumor cells reach and invade leptomeninges in various ways, such as through hematogenous spread, perineural and perivascular lymphatic spread, and direct seeding from the brain parenchyma.[20, 21] Direct spread to the CSF space is particularly relevant to primary brain tumors,[22] as tumor contact with the subventricular zone is a known risk factor for LMD.[23] Moreover, tumor cells may disseminate through the CSF with a predilection to regions with slow CSF flow or gravity-dependent sites, such as the basal cisterns or posterior fossa.[24] In this study, the majority (85.7%) of initial tumors were located in the periventricular zone, and no basal cistern opening was observed during the surgical procedures in our series. Therefore, we hypothesized that ITD also develops due to the direct spread of tumor cells to the CSF spaces. The timing of tumor cell spreading is unknown; it may constantly occur during tumor growth or occur at the ventricle opening by surgical intervention. The tumor cells drift from the supratentorial region to the posterior fossa via the lateral and third ventricles, the aqueduct, and the fourth ventricle. They depart from the fourth ventricle through the median and lateral apertures, reach the extra-cerebellar space, and eventually infiltrate the cerebellum and brainstem. Thus, the brainstem infiltration reported by Drumm et al. is considered a terminal stage of GBM.
Previous studies have identified that other than a tumor located in the periventricular zone, the risk factors for LMD are long survival [9], young age [7, 9], larger initial tumor size [7], and MGMT promoter methylation [6]. Patients with ITD have a longer OS (19.7 months) than the general GBM population (10.1–15.2 months) [25–29]. By contrast, the median age of the patients (66.0 years) is consistent with that of the patients with GBM (63–65 years) [27, 30–32]. Moreover, the MGMT promoter's low methylation status (21.4%) in the patients was lower than that in the general GBM population (45%) [33]. and the extent of resection (total resection 28.6%) did not exceed previous reports (20–43%) [30, 34]. The initial tumor size of ITD (mean volume, 26.5 cm3) is comparably smaller than that in a previous study (33.2 cm3) [35]. Hence, ITD may be an atypical subtype of LMD with a unique biological nature.
The median patient survival after ITD was 7.1 months, which is better than that of the patients with LMD (2.1–5.7 months) [4–8]. One clinical advantage of ITD is that radiotherapy would be a good therapeutic strategy. In the case of supratentorial GBM recurrence, including dissemination, radiotherapy is not a widely approved therapeutic option since a total of 40–60 Gy of maximal dose radiotherapy has already been performed in the supratentorial area as an initial treatment. Therefore, therapeutic options for LMD are generally limited to chemotherapy such as bevacizumab and lomustine therapy [36]. By contrast, in the case of ITD, the infratentorial region is generally a radio-intact area, and additional radiotherapy is a good therapeutic strategy. No optimal radiation dose regimen has been established for ITDs. Here, three different doses (25, 36, and 60 Gy) of radiotherapy were administered, and no significant difference was observed in their prognoses. A previous study also used various doses ranging from 21 Gy to 45 Gy for radiotherapy [13]. The optimal radiation dose regimen for ITD should be established in future studies. At the same time, choosing the optimal radiation dose regimen for ITD patients based on their age and physical status is equally important.
One finding was the efficacy of dabrafenib (BRAF inhibitor) plus trametinib (MEK inhibitor) for ITD in a patient with the BRAF-V600E mutation. The incidence of BRAF-V600E mutation in adult GBM is approximately 1–3% [37, 38]. Due to its rarity, the effectiveness of BRAF inhibitors in GBM is unclear [39]. Burger et al. have reported one case of ITD of BRAF-V600E mutation and IDH-wildtype GBM [40]. In their study, one month of dabrafenib treatment resulted in a nearly complete response to ITD. In this study, dabrafenib plus trametinib was administered as a second-line chemotherapeutic agent to patient 4, whose tumor harbored the BRAF-V600E mutation. This patient experienced a favorable therapeutic response and extended survival after ITD diagnosis (15 months). Thus, dabrafenib plus trametinib may be a promising therapeutic strategy for the BRAF-V600E mutation and IDH-wildtype in patients with GBM, even after ITD development.
The expected survival improves for patients who develop ITD earlier at the age of 50 years or those with a high KPS (> 70). The age of 50 years and KPS of 70 are essential thresholds for patients with GBM. Li et al. have reported that age (< 50 vs. ≥ 50 years) and KPS 70 produces a significant split and set as the bifurcation of the recursive partitioning analysis model of GBM [41]. In the general population with GBM, younger age and good performance status may be beneficial for patients with ITD.
The symptoms observed in ITD were often systemic, such as gait disturbance and nausea, rather than neurologically focused symptoms such as hemiparesis or language deficits. These symptoms can easily be misdiagnosed as side effects of chemotherapy and can be observed on the left side. In fact, four cases (57.1%) exhibited new symptoms several weeks before MRI examination. Thus, ITD must be considered when patients present with unusual and neurologically nonfocused symptoms.
Moreover, patients with ITD often experience persistent nausea and intractable vomiting, leading to appetite loss. Cohen et al. have reported three cases of uncontrollable vomiting from a GBM that disseminated to the fourth ventricle [13]. In these cases, additional irradiation to the infratentorial region achieved complete remission of symptoms. Here, patients 1 and 2, who developed ITD in the fourth ventricle, experienced persistent nausea. Although the serotonin 5-HT3 receptor antagonist did not affect the symptoms, chemotherapy and radiotherapy relieved intractable vomiting within one week, as previously reported.
By contrast, no improvement was observed in gait disturbance, which makes it challenging for patients to independently perform activities of daily living. Consequently, the patients became weaker, and their performance status declined. Even young patients hardly maintained their KPS score after treatment. Their clinical course suggests that once the physical performance of ITD drops, there is no optimal treatment to improve it. Early detection and intervention are the best efforts to improve the performance status and prognoses of patients with ITD.
The main limitation of this study was the small sample size due to the rarity of ITD; therefore, our results need to be carefully interpreted. Another significant limitation is that the therapeutic strategies employed may be biased based on patient performance status. Since this was a retrospective study, patients with a good performance status might have received more intensive treatment, and those with a poor performance status might have undergone more palliative treatment. These therapeutic differences reflect realistic clinical decisions, although they hinder the objective assessment of outcomes in patients with ITD.