The safety profile of Tumor Treating Fields (TTFields) therapy in glioblastoma patients with ventriculoperitoneal shunts

Glioblastoma (GBM) is the most aggressive primary malignant brain tumor in adults, with a high recurrence rate, an extremely poor prognosis, and an estimated 5-year survival rate of 6–22% [1]. Until recently, the optimal treatment for newly diagnosed GBM (ndGBM) has been maximum safe resection, followed by radiation with concomitant temozolomide (TMZ) [2]. However, the addition of Tumor Treating Fields (TTFields; Optune^®, Novocure^® GmbH, device manufacturer) therapy to maintenance TMZ has now been incorporated into the ndGBM treatment paradigm, following its approval in the European Union (EU), United States (US), Japan, and China [3‐7]. Furthermore, TTFields therapy has a category 1 recommendation for adult patients with ndGBM in the National Comprehensive Cancer Network guidelines [8, 9]. TTFields therapy is also approved in the US and EU, as well as other countries for the treatment of adult patients with recurrent GBM (rGBM) [5] and for malignant pleural mesothelioma (MPM) [10].

TTFields therapy is a first in class, noninvasive, locoregional cancer treatment delivered via a portable medical device that is designed to be integrated into daily life, while maintaining patients’ quality of life [5, 10‐12]. TTFields work by exerting electric forces on polar components within cells, disrupting their normal localization and function, and selectively act on cancer cells due to their distinct characteristics, including rapid proliferation, morphology, and electrical properties, without significantly affecting healthy cells or tissue [13‐18].

Approval for use in ndGBM was based on a phase 3 clinical study (EF-14; NCT00916409) that demonstrated significantly improved progression-free survival (PFS) for TTFields therapy concomitant with TMZ (6.7 months) versus TMZ alone (4.0 months) and a significantly improved overall survival (OS) for TTFields therapy concomitant with TMZ (20.9 months) versus TMZ alone (16.0 months) [19]. Additionally, 5-year OS rates were more than double that of TMZ alone (5% vs. 13%; P = 0.04) [19]. Improvements in outcomes were also observed in the EF-11 study, in which TTFields therapy was compared to the best standard of care in patients with rGBM [20]. Furthermore, clinical efficacy has also been demonstrated in a range of other solid tumors, including non-small cell lung cancer, liver, ovarian, and pancreatic cancer, when used concomitantly with systemic therapies and alongside radiation [21‐26]. In terms of safety, clinical and real-world data demonstrate that TTFields therapy has a favorable safety profile, characterized by an increased rate of dermatologic adverse events (AEs), but a low rate of systemic AEs compared with chemotherapeutic regimens [19, 20, 27, 28].

Up to 10% of patients with GBM may develop hydrocephalus, for which ventriculostomies or ventriculoperitoneal (VP) shunts may be needed [29]. There are limited safety data on the use of TTFields therapy with devices such as programmable VP shunts. Therefore, further investigation of the safety and feasibility of TTFields therapy in patients with GBM requiring VP shunts may provide rationale to provide access to TTFields therapy particularly for those in this vulnerable population.

Here, we report the results of a retrospective analysis of unsolicited post-marketing surveillance data to assess the safety of TTFields therapy in adult patients with ndGBM or rGBM with a VP shunt.

This global post-marketing surveillance study provides the first real-world safety data for TTFields (200 kHz) therapy in highly burdened patients with GBM, hydrocephalus, and a surgically implanted VP shunt. The VP shunt population was representative of the real-world GBM population in terms of male:female ratio and average age [30, 31].

The Stupp protocol was published in 2005, making maximal safe surgical resection, radiation therapy, and TMZ the cornerstone of ndGBM treatment. [2]. In 2017, Stupp et al. reported on the outcomes of the EF-14 study, demonstrating that the addition of TTFields therapy to the Stupp protocol led to significantly improved outcomes (OS and PFS) [32]; as a result, TTFields therapy became part of the treatment regime in ndGBM and rGBM. Alternative treatment options, in particular in rGBM include chemotherapeutic agents such as bevacizumab, and radiotherapy. Treatment options in the experimental spectrum include immunotherapy and targeted therapies, which are used in conjunction with surgery and/or radiotherapy. As a result of such advances, patients are now living longer, which has led to the emergence of long-term complications of the disease and treatment [33], such as disturbances to the cerebrospinal fluid circulation, leading to clinical and symptomatic hydrocephalus. Neurological deterioration associated with the development of hydrocephalus has been observed in 3–15% of patients with GBM [29, 34‐36]. Such complications significantly reduce a patient’s quality of life [29, 34‐36]. By the time hydrocephalus develops, patients have generally already undergone surgery, radiation therapy, chemotherapy, long-term steroid treatment, immunotherapy, etc., increasing their vulnerability to treatment-related AEs and further complications. Although surgical shunts are often recommended to restore and maintain cerebrospinal fluid levels, significantly improving symptoms, functional performance, and quality of life, they rarely impact survival rates, therefore, treatment is needed to combat neurological deterioration [37‐39]. Many of the approved treatments for GBM are associated with significant systemic side effects that can have a detrimental impact on patient quality of life and may be of limited benefit in patients with multiple comorbidities, such as those with VP shunts [40]. VP shunts are foreign bodies, that carry an inherent risk of infection and as such, patients harboring VP shunts are more susceptible to shunt site infections [41]. Safety data on appropriate GBM treatments for patients with VP shunts are currently lacking. Therefore, it is important that the safety and tolerability of GBM treatments in patients with implanted shunts should be thoroughly assessed.

To assess the overall safety profile of TTFields therapy in patients with VP shunts, it is important to distinguish between AEs that are associated with the shunt rather than with TTFields therapy.

No TTFields therapy-related systemic AEs or TTFields therapy-related shunt dysfunctions were reported.

In fact, the incidence, nature, and severity of AEs in this population, regardless of GBM disease status (ndGBM or rGBM), were very similar to those observed in a non-shunt GBM population. Furthermore, the number of seizures and reports of hydrocephalus were in line with that expected for patients with GBM.

Results presented here are in line with those reported in prior TTFields therapy studies in GBM patients without shunts, including the phase 3 EF-14 (ndGBM) [19] and EF-11 (rGBM) [20] clinical studies, PRIDE registry (rGBM) [27], and the global post-marketing surveillance data analysis (> 10,000 patients with GBM) [28]. The lack of any new safety signals is encouraging given the vulnerable nature of patients with GBM and surgically implanted VP shunts. The safety profile of TTFields therapy is further supported by data from a recent subgroup analysis of elderly patients from the EF-14 study, another vulnerable population [42].

Indeed, skin AEs associated with TTFields therapy were mild-to-moderate skin irritation and can typically be managed by early prophylactic interventions and good patient management strategies, including optimal shaving and shifting the array position (~ 2 cm) or by the use of topical corticosteroids or antibiotics [5, 43]. The irritation reported here generally resolved after a brief pause and did not require any substantial break in treatment. Although, 22 patients described experiencing a warm/heat sensation, these events were typically attributed to inadequate adherence of the array to a patient’s scalp. It is important to note that the device delivers TTFields has a protective sensor-based shut-off feature if the temperature rises.

Although data on shunt complications in patients with GBM treated with TTFields therapy in the presence of VP shunts are limited, there is one publication reporting on the case of a patient with a programmable shunt who received TTFields therapy, which showed that shunt valve settings were stable over the five days during which the patient received TTFields therapy [44]. In addition, there are some studies that have reported a much higher incidence of shunt-complications than those identified in the present analysis. One study analyzed data from 62 patients with supratentorial glioma and VP shunts, of whom 41 had GBM. Among these patients, 27% had complications related to VP shunts [39]. Another study showed that eight of 16 patients (50%) with GBM and a VP shunt had experienced shunt-related complications, with three patients dying as a result [38]. A further study reported that shunt complications required surgical revision in four of 12 patients (33%) with high-grade glioma who had either VP or cystoperitoneal shunts [45]. These data provide a baseline for expected AEs in patients with GBM and a VP shunt. Considering this, our analysis identified 15 AEs from 14 patients (9%) that were VP shunt-associated and TTFields therapy-related AEs. Furthermore, only five (3.2%) shunt-associated, TTFields therapy-related events were identified in this patient population, all of which were dermatological complications at the scar site, associated with array placement. These findings are in line with previous clinical data that show an association between TTFields therapy and skin AEs.

The retrospective nature of this study represents a limitation as the analyses could not be statistically powered meaning that comparative statements should be regarded as observational only. As analyses were retrospective and not actively solicited, a full medical history and details related to AEs were not available for all patients – missing information could potentially have an impact on interpretation of the study results. Furthermore, an inherent limitation of observational and retrospective studies is that there is no control over prior therapies received. In this case, information on treatments used prior to and concomitantly with TTFields therapy (for example, steroids or anti-cancer treatments such as bevacizumab), were not included as this information was not available for all of the patients. The impact of these treatments on safety outcomes cannot be adjusted for and should also be considered when evaluating findings reported here, since some therapies may have affected the incidence of reported AEs; for example, steroid use and bevacizumab can increase skin fragility. Without information on prior therapies received, it is difficult to accurately assess the relatedness of AEs to TTFields therapy. Furthermore, AEs were not graded for severity as per the protocol, unlike data collected in a controlled clinical trial setting, which may have also impacted the occurrence of TTFields therapy-related AEs. Finally, as the study was retrospective, patients could not be followed up for subsequent safety outcomes.

TTFields therapy employs electric fields in a frequency range of 100 kHz to 500 kHz, which is too high to stimulate tissue and too low to have ionizing or significant heating effects [13, 46]. TTFields therapy is delivered at a specific frequency based on the cancer cell type being targeted, allowing TTFields to enter cells more effectively [47]. Of note, TTFields are not electro-magnetic fields but electric fields and, although there is a magnetic field that results from applying TTFields, it is low-level and not expected to have any relevant impact on magnetic adjustable (programmable) valves. Given that some VP shunts operate based on a magnetic system [48], there has been concern that concomitant use of TTFields therapy would impact normal function. Nevertheless, in the interests of safety, as it is not possible to control all conditions that could theoretically impact the function of a particular shunt, patients with programmable shunts have typically been excluded from previous studies. The findings reported here suggest that the use of TTFields therapy in patients with VP shunts is feasible; prospective data would be of additional value in this patient population.

Our analysis of 156 patients with GBM and implanted VP shunt for the relief of hydrocephalus provides evidence that TTFields therapy is feasible and well-tolerated and does not seem to interfere with the normal function or the effectiveness of the VP shunt. Furthermore, based on the case study presented here, high usage can be achieved in this patient population, which may improve efficacy. In the absence of large-scale randomized controlled trials, these real-world observational data, supportive of previous clinical and real-world evaluations of TTFields therapy in patients with GBM and across varied solid tumor types, provide insights into the potential role of TTFields therapy in patients with VP shunts. These findings, together with further investigations and ongoing clinical experience of TTFields therapy use in patients with GBM, will hopefully contribute to improved decision making and patient counselling in terms of treatment options, with the aim of making TTFields therapy available to patients with VP shunts, addressing the need in this population.

Based on presented post-marketing safety surveillance data spanning almost a decade, the use of TTFields (200 kHz) therapy in adult patients with GBM and hydrocephalus requiring a VP shunt demonstrated feasibility with a tolerable safety profile. Overall, these data provide further evidence of the broad applicability of TTFields therapy. Commonly reported AEs were localized, manageable, beneath-array skin AEs. The data are supportive of previous phase 3, registry, and post-marketing studies of TTFields therapy in GBM and across varied solid tumor types, with no new safety signals or added safety concerns observed in this equitable patient sample size (> 150 patients). Moreover, the lack of observed effect of TTFields therapy on shunt effectiveness indicates sustained shunt functionality in the management of hydrocephalus symptomatology.

The safety evidence presented here suggests that the use of TTFields (200 kHz) therapy in the presence of VP shunts is an efficacious and viable treatment method in patients with GBM and may help address an important unmet medical need in this heavily burdened patient population with limited treatment options.