The role of GammaTile in the treatment of brain tumors: a technical and clinical overview

GammaTiles are bioresorbable, conformable, 20 × 20 × 4 mm collagen squares containing four radioactive titanium-encapsulated cesium-131 (Cs-131) seeds per tile (Fig. 1). Upon completion of resection, the neurosurgeon lines the tumor cavity with sufficient tiles to cover the surfaces at risk for recurrence. This permanently implanted device functions as both a seed carrier and three-dimensional spacer, offsetting seeds 3 mm from the tumor cavity surface and 10 mm from each other (Fig. 2). The 10 mm fixed inter-seed spacing aims to prevent dosimetric “hot spots” and “cold spots” that could arise from irregular spacing of seeds and can occur with more traditional brain BT methods, such as stranded seeds [19]. The 3 mm offset between seeds and the tile surface prevents the very high doses of radiation that would occur if seeds were placed directly on brain.

The radioactive Cs-131 seeds used in GammaTiles are manufactured to have a fixed source strength of 3.5 U (cGy cm² h⁻¹) on day of implant. The GammaTile design and source strength results in a physical dose of approximately 60 Gy to a depth of 5 mm in brain (Fig. 3); however, cavity shape, cavity size, and number of tiles can affect specific implant dose distributions. Compared to more traditional brain BT using seeds with a range of source strengths and heterogeneity of prescription doses [21, 22], standardization of source strength at implant reduces dosimetric variability across patients and may simplify studying outcomes.

Cs-131 has a half-life of 9.7 days, compared to 59.4 days for iodine-125, which has also been used in brain BT. The relatively short half-life of Cs-131 ensures the planned dose is given in a shorter period at a higher initial dose rate, which may better address highly proliferative malignant brain tumors [23]. The short half-life also helps ensure the planned dose is delivered before collagen tile reabsorption and seed displacement [24‐27].

GammaTile therapy requires a multidisciplinary care team approach [28]. Neurosurgery, neuro-oncology, and radiation oncology teams collaborate on patient selection and accurate case planning. Based on a preoperative MRI, the surface area of the expected postoperative tumor cavity deemed at risk for recurrence is estimated, considering portions of the cavity that may not need tiles (e.g., surgical tract) and potential intraoperative cavity contraction. After tumor resection, the tiles are placed on the tumor bed with the 3 mm offset from the seed facing brain parenchyma to achieve the desired seed-to-brain offset.

The collagen component of the GammaTile is the same material neurosurgeons have used to reconstruct dura for decades, and the ease of usability is demonstrated with an average implantation time of 2–5 min [24]. The fast implant time helps minimize radiation exposure to perioperative staff and limits the time patients spend under anesthesia. In a study of 22 patients treated with GammaTile (including two patients who had multiple GammaTile treatments for separate tumors), measured and modeled radiation exposure for healthcare workers and caregivers was below regulatory limits for medical personnel and the public [29].

Following surgical closure, a postoperative radiation survey measurement is performed to quantify the maximum radiation exposure one meter from the implant site to verify the patient meets regulatory limits for patient release. Using U.S. Nuclear Regulatory Commission recommendations, patients may be released based on 1-meter measurements of ≤ 6 mrem/hr [30]. In a single institution study, the reported mean survey measurement was 1.83 mrem/hr with a range of 0.5–3.5 mrem/hr for 13 patients [31]. In addition to the standard postoperative MRI exam assessing extent of resection, a thin-cut non-contrast brain CT is obtained, typically 12–72 h postimplant. Like other forms of permanent seed BT, the CT is used to document the position and number of implanted sources as well as the isodose lines/radiation field (Fig. 4) [32]. Importantly, the outcome data (LC, toxicity) described below results from surgically targeting the oncologic at-risk tissue with GammaTiles rather than trying to meet pre-specified dosimetry parameters. To date, no specific postimplant dosimetric parameters have been correlated with probability of LC or toxicity in the case of recurrent GBM [33] or brain metastases [8]. Predictive postimplant dosimetric parameters may become apparent as trial data matures (Table 1).

Brain metastases are the most common malignant brain tumors in adults, and they pose a burden for patients with cancer both at initial diagnosis and re-occurrence. The use of SRT for intact, typically smaller, non-resected brain metastases confers LC on the order of ~ 85% [34]. Durable LC in the postoperative setting for larger brain metastases requiring resection has been more challenging, with LC at 60–79% [1‐3]. Whole brain radiotherapy is another option, but comes at the cost of neurocognitive toxicity and still results in a local failure rate of ~ 14% when used in the postoperative setting [1].

Intraoperative treatment with GammaTile implantation may mitigate factors that lead to worse LC after resection. Changes in the resection cavity between SRT planning, MRI, and SRT delivery could lead to a marginal miss, especially due to the highly conformal nature of SRT [7, 9]. Also, time delay from surgery to SRT initiation (which can result from a skilled nursing stay, coordination with systemic therapies, and logistics for setting up radiation oncology consultation, CT simulation, and radiation delivery [35]) leads to worse LC, with a local failure rate of 2.3% if SRT starts before 4 weeks, 14.5% between 4 and 8 weeks, and 48.5% if started after 8 weeks postoperation [35]. Some patients are lost to follow up and never receive SRT, with one prospective trial of post-operative SRS demonstrating 22% of patients never received the planned SRT course [36]. Furthermore, a subset of aggressive brain metastases can rapidly recur after complete resection but before postoperative SRT, even if SRT is delivered before 4 weeks [8]. Compared to postoperative SRT, GammaTiles are implanted at resection, guaranteeing adjuvant radiation starts immediately.

GBM is the most common malignant primary brain tumor in adults. Most re-occur within 2 cm of the resection cavity [4, 24, 47‐49]. Despite great efforts over the last ~ 20 years to improve the standard of care for the treatment of newly diagnosed GBM, median OS from diagnosis continues to be poor, around 16–20 months [50].

At recurrence, there is no established standard of care, and available options include surgical resection, re-irradiation, systemic therapies and/or immunotherapies, alone or in combination [12‐14, 50‐52]. None of the available treatment strategies have shown significant OS via randomized trials.

With any new technology in oncology, the development of mature data takes significant time. Multiple clinical trials are evaluating the efficacy and safety of GammaTile. Pending the completion of these trials, and 5 years after FDA clearance, we have reviewed the accumulated published data across multiple tumor types.

GammaTile is one method of utilizing Cs-131 brachytherapy sources for brain tumors. The other contemporary method that has been reported upon is implantation of vicryl sutures containing Cs-131 seed sources, without the 3-dimensional collagen tile carrier/spacer function. This is often referred to as “stranded seeds.” Table 2 lists the institutional experiences with Cs-131 that include local control and toxicity data that utilized either GammaTile or stranded seeds.

These data suggest LC outcomes with surgery and Cs-131 use for brain metastases, GBM, and meningioma seem promising, particularly in the context of other strategies currently employed for these complicated clinical scenarios and tumor types. In addition, the rate of surgical or radiation AEs after GammaTile use is similar to that of surgery alone or other forms of radiation. Taken together, GammaTile appears to be a useful adjuvant for resected brain tumors.