Chasing a rarity: a retrospective single-center evaluation of prognostic factors in primary gliosarcoma

Gliosarcoma (GS) is a rare variant of glioblastoma multiforme (GBM), accounting for about 2% of cases [1‐3]. It can be divided into primary (de novo) and secondary (after prior GBM treated with radiation) GS [4, 5]. It is typically diagnosed in the fifth or sixth decade and is about twice as common in men [2, 5, 6]. GS is usually located in the supratentorial region, with a predilection for the temporal lobe [5, 7, 8]. Presenting symptoms include signs of raised intracranial pressure (e.g., headaches, nausea, and vomiting), visual disturbances, or seizures [5, 9]. Histopathologically, it is characterized by a biphasic growth pattern, including both a glial and an atypical sarcomatous component, descending from a monoclonal origin [10]. In the latest WHO classification, it is regarded as a subtype of the isocitrate dehydrogenase (IDH) wildtype GBM, although IDH-mutated GS has been described [1, 11].

Because of its only sporadic occurrence, GS is treated in a similar manner to classical GBM, with a multimodal therapeutic approach including maximal safe surgical resection, external beam radiotherapy, and temozolomide-based chemotherapy [12‐14]. A historic cohort showed a median overall survival (OS) of only 4 months if left untreated [15]. With treatment, GS continues to have a poor prognosis, with a median OS around 15 months comparable to that of GBM, thus making it one of the tumors of the central nervous system with the lowest relative survival rate [1, 7, 16‐18].

In the current study, we present a cohort of 26 GS patients treated at a single center, reviewing their demographic, clinical, and treatment characteristics (16 distinct factors in total). Conclusively, we aim to identify possible independent factors that might be related to an improved outcome.

In this retrospective analysis, we present a cohort of 26 GS cases treated at a single center. The demographic and clinical characteristics of our cohort are comparable to previously conducted research [2, 5‐7, 9].

Patients presenting with a single symptom upon diagnosis showed improved survival. In 3 of these 6 cases, the symptom was a new-onset seizure in an otherwise healthy individual, warranting prompt investigation, and thus quickly leading to diagnosis and treatment of an intracranial tumor in an early stage. These patients also generally showed a higher Karnofsky performance status.

The extent of tumor resection is a well-established independent prognostic factor for improved OS in GBM [19]. It has also been described in GS [2, 18, 20, 21, 32]. In the current study, GTR did not lead to improved survival. We reason that this might be attributed to the small sample size, since all studies reporting a significant influence on survival had larger patient numbers.

Whether MGMT promotor methylation status influences survival in GS remains a matter of ongoing debate. Hypermethylation leads to epigenetic silencing of the MGMT gene. Its product, a DNA repair enzyme, restores alkylating agent-induced damage [22]. Hypermethylation of the MGMT promotor thus reflects the efficacy of alkylating agents such as temozolomide, and has been thoroughly associated with improved outcome in GBM patients [19]. Reports on the prevalence of MGMT promotor hypermethylation and its association with survival in GS are conflicting, however. Some authors reported less frequent hypermethylation in comparison to GBM and used this as an explanation for their observed worse prognosis in GS [7, 23, 32]. In contrast, others suggested more hypermethylation in GS [21, 22]. In the current study, MGMT promotor methylation status did not influence survival, consistent with some previous reports [9, 24]. The largest GS registry study to date (with 1102 included patients) found no apparent difference in MGMT promotor hypermethylation between GS and GBM, nor did it influence survival in GS patients [18].

Adjuvant temozolomide therapy has become the standard of care in GBM management, leading to a statistically significant and clinically relevant survival benefit [12]. This has also been reported to some degree in GS, albeit mostly in studies with small patient numbers [6, 20, 33, 34]. The largest registry study so far, however, confirmed the superiority of trimodality therapy also in GS [18]. In the current study, there was no statistically significant difference in survival among patients receiving adjuvant temozolomide therapy.

A multitude of other favorable prognostic factors has been reported in the literature, including younger age at diagnosis [2, 18, 23, 25‐27, 35], female sex [18], temporal tumor location [7], preoperative performance status [23, 25], and tumor size [23, 26]. These are, however, often not replicated because of small patient numbers, and were also not observed in the present cohort.

The median OS in this cohort was 12 months. There is wide variation in reported median OS in the available literature (Fig. 3; range 5.7–18.5 months). The largest conducted registry study on GS reported a median OS of 10.7 months, comparable to the current cohort [18].

Our study harbors several limitations, most importantly its retrospective nature and relatively small sample size. Furthermore, we did not compare our data to a control group of GBM patients. In earlier cases, information on survival was often missing, which could have been a potential source of bias. However, because of the rarity of GS, we reason that an accurate depiction of GS patients treated at our center is of value and adds to the existing data on GS management.

With a 5-year OS rate of only 5.6%, GS remains a rare tumor entity with a dismal prognosis [17]. Although overlap with GBM exists, GS appears to be a separate tumor entity, with variation in reported prognostic factors. Prospective studies with data on molecular pathology are needed to unequivocally establish such patient and tumor characteristics, aiding patient management and improving care for all GS patients. National and international collaborations could serve useful herein, to facilitate patient recruitment. Until then, a multimodality approach, with surgery aiming for complete resection followed by radiation and chemotherapy, seems to be the most favorable option.