Abstract 3450: Impact of model inaccuracy on dose estimation in TTFields therapy

Introduction: Tumor Treating Fields (TTFields) is a modality for treating Glioblastoma (GBM). A recent study combining clinical data and simulations demonstrated that the simulation-based dose estimation at the tumor level is directly correlated with patient survival (Ballo M, et al. Int J Radiat Oncol Biol Phys. 2019;104(5):1106-1113). This study highlights the importance of computer-modelling based TTFields treatment planning in the clinic. Performing such planning in a meaningful manner requires an understanding about how inaccuracies in the patient model influence calculated TTFields distributions. In this study we show the effect of local perturbations in the model, errors in tumor segmentation, and inaccuracy in the patient head model on TTFields dose estimation.

Methods: Computational studies were performed with Sim4Life software. 1) Local model uncertainty outside of tumor region: To create defects in the models, conductive spheres with varying conductivities and radii were placed into the model’s brains at different distances from the tumor. Virtual transducer arrays were placed on the models, and delivery of TTFields numerically simulated. The error in the electric field induced by the defects as a function of defect conductivity, radius, and distance to tumor was investigated. 2) Tumor segmentation uncertainty: For illustrating segmentation errors in tumors and necrotic cores, layered spheres with various conductivities and radii were placed in different head locations. Tumor conductivity uncertainty was defined as (σ-σref)/σref where σref = 0.24 S/m is the standard tumor conductivity and σ is the varied conductivity over the tumor. The normalized tumor radius was calculated when each of the varied radii was normalized to the results obtained by all the other combinations.

Results: The results show that that when a defect of radius R is placed at a distance, d, from the tumor that is larger than seven times R, the error is below 1% regardless of the defect conductivity.

Conclusions: Our models and results show the impact of uncertainty in segmentation of the tumor or other tissue distant from it on the TTFields dose estimation. These results could serve as a guideline for model creation and tissue segmentation that would lead to optimal dose estimation in TTFields treatment planning.

Citation Format: Tal Marciano, Oshrit Zeevi, Zeev Bomzon. Impact of model inaccuracy on dose estimation in TTFields therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3450.

EXTH-75. APPLICATION OF TUMOR TREATING FIELDS (TTFIELDS) TO THE HEAD AND TORSO OF MICE WITH THE DEDICATED INOVIVO SYSTEM

INTRODUCTION

Tumor Treating Fields (TTFields) therapy is an approved anti-cancer treatment for glioblastoma and mesothelioma. TTFields are delivered to patients continuously by two sets of arrays placed on opposite sides of the body at the tumor region to generate two perpendicular electric fields. Previously, in vivo studies of TTFields were limited due to the lack of a dedicated system that could maintain continuous and adequate contact of the arrays with the animal’s skin as well as the stress imposed on the animals by individual housing and the motility limitations they experience during treatment.

METHODS

Different electrode layouts were explored to optimize the intensity of the electric fields delivered to the target locations (therapeutic threshold >1 V/cm). The ability of various adhesive materials and wire coiling prevention strategies to increase TTFields device usage was examined. Stress reduction with different housing methods was evaluated via clinical examination of the animals.

RESULTS

Optimal array layouts were identified based on simulation data for TTFields delivery to the torso or the head of the mouse. Compacting conductors into a single printed circuit cable connected to a novel electric swivel machine resulted in fewer wire entanglements, and the improved adhesives resulted in fewer array replacements, overall elevating device usage. Improved cage design permitted pairs of mice to maintain social interactions while individually housed. Less weight loss was seen for animals housed in the dyadic relative to the standard solitary cages, indicating reduced stress.

CONCLUSIONS

The inovivo system provides means for continuous delivery of therapeutic levels of TTFields to the head and torso of mice while minimizing animal stress and increasing device usage. The new head arrays enable application of TTFields to the head of mice for the first time, allowing expansion of glioblastoma treatment research.

A computational study of the relation between the power density in the tumor and the maximum temperature in the scalp during Tumor Treating Fields (TTFields) therapy

In this work we investigated the relation between the power density in the tumor and the maximum temperature reached in the scalp during TTFields treatment for glioblastoma. We used a realistic head model to perform the simulations in COMSOL Multiphysics and we solved Pennes' equation to obtain the temperature distribution. Our results indicate that there might be a linear relation between these two quantities and that TTFields are safe from a thermal point of view.

Abstract 1435: The distribution of Tumor Treating Fields is affected by cell confluence and pores in the membrane

Introduction: Better understanding of the interaction between Tumor Treating Fields (TTFields) and living cells would help to elaborate on its mechanism of action. Numerical simulations investigating the electric field distribution within isolated cells have shown that during metaphase a uniform electric field forms within the rounded cells, and during cytokinesis a non-uniform field forms at the furrow, leading to strong dielectrophoretic (DEP) forces that can disrupt cell division. However, preclinical studies have shown that TTFields influence cells during earlier stages of cell division. Hence, field non-uniformity within the cell during cytokinesis cannot provide a full explanation for how TTFields exerts an anti-mitotic effect on cells. We hypothesized that strong DEP forces could arise at the boundaries between cells and in pores on the membrane when applying TTFields to cell culture. We tested this hypothesis by using numerical simulations to investigate how the clustering of cells and pores within the membrane influence TTFields distribution.

Methods: COMSOL was used to numerically simulate delivery of TTFields to clusters of round cells placed in a hexagonal arrangement. The influence of the distance between the cells on field distribution was investigated. The effect of pores in the cell membrane on field distribution was also investigated. A generic analytical model was developed for predicting conditions for field amplification in the TTFields frequencies.

Results: Placing round cells in clusters separated by ~10 nm resulted in regions of highly non-uniform fields within the cells and very strong DEP forces at the intercellular level. Maximum field intensities between cells were observed at 200 kHz. Very strong gradients in the electric field were observed around pores placed in the membrane, and a strong field amplification was observed at 200 kHz regardless of the pore size. For a 10 nm pore, up to 450 V/cm is generated at the pore's vicinity. 100 nm pores generate more than 130 V/cm for an applied external field of 1 V/cm. Electroporation generates enough energy for overcoming the internal KT energy at pores that are ≤30 nm and may allow for dipole alignment.

Conclusions: Cells in close proximity to one another creates gradients in the electric field, and are associated with strong DEP forces that enhance the effects of TTFields on cells. Strong DEP forces in the membrane may provide a physical mechanism by which TTFields enhance membrane permeability. The strong field amplification could cause local heating, particle diffusion, and impact the cytoskeleton in the vicinity of membrane pores. These simulations have validated a generic model that predicts the conditions for strong field amplifications in cells at TTFields frequencies, allowing further elaboration of the mechanism of action. Our model fits well with other analytical models; the furrow model, cells in confluence, or electroporation.

Citation Format: Tal Marciano, Shay Levi, Eduard Fedorov, Zeev Bomzon. The distribution of Tumor Treating Fields is affected by cell confluence and pores in the membrane [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1435.

Abstract 3070: Lung cancer TTFields treatment planning sensitivity to errors in torso segmentation

Introduction: Tumor Treating Fields (TTFields) is a treatment modality for glioblastoma multiforme (GBM) and other malignant tumors, utilizing low-intensity, intermediate frequency (100-300kHz) alternating electric field delivered through noninvasive transducer arrays. TTFields are currently being tested in a phase III clinical trial for the treatment of advanced Non-Small Cell Lung Cancer (NSCLC). Previous studies have demonstrated that the treatment efficacy increases when the dose of TTFields delivered to the tumor is at least 1 V/cm. Thus, personalized simulations to estimate the dose are performed as part of the patient treatment planning. For NSCLC treatment, these simulations require the segmentation of torso tissues. We are currently developing a semi-automatic software for the segmentation of these tissues. Achieving highly accurate segmentation is time consuming and not always necessary. Therefore, in this study we assess the level of segmentation accuracy that is required for TTFields dose estimation.

Methods: We introduce a novel pipeline for evaluating the sensitivity of TTFields treatment planning errors in the segmentation of torso tissues. To this end, we incorporated an expert-segmented torso volume and artificially placed NSCLC tumors of different sizes in the vicinity of the tested torso tissue. We employed relaxations to the segmentation algorithm and derived the TTFields distribution with and without the simulated segmentation errors. We then computed the treatment planning error as the per-voxel average deviation of the TTFields dose computed over the erroneous segmentation from the reference one. The dose was calculated over the simulated tumor location and over the entire lung volume.

Results: We have evaluated our sensitivity analysis pipeline on 9 different test cases. Based on previous analysis, we have set our tolerable target error to be 0.1 V/cm. Our results show that 6 out of the 9 tests were within the target, whereas the other 3 exceeded the target, requiring algorithmic refinements.

Conclusions: We have presented a method for estimating the sensitivity of lung cancer TTFields treatment planning to segmentation errors. This method is used to evaluate the required accuracy of the semi-automatic torso segmentation tool that we are currently developing. Our method can be further extended to other types of tumors in different regions of the body.

Citation Format: Hadas Ben Atya, Oren Peles, Reuben Shamir, Zeev Bomzon. Lung cancer TTFields treatment planning sensitivity to errors in torso segmentation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 3070.

Abstract 3071: A method for infratentorial structures segmentation for tumor treating fields treatment planning

Introduction: Tumor treating fields (TTFields) is an FDA approved treatment for the management of glioblastoma multiform (GBM) and mesothelioma (MPM) and is associated with a significant extension to patients' survival. We have developed software to optimize the location of the TTFields transducer arrays (TAs) on GBM patients' head to maximize the field in the tumor region. To that end, we first segment the patients' head MRIs into five normal and three abnormal tissues. However, the current method for head segmentation was found to be less accurate at the infratentorial regions, mark some sinuses with cerebrospinal fluid (CSF), and does not identify the cerebellum and brain-stem. Therefore, it is assuming the infratentorial structures have the same electrical properties as in the cerebrum.

Methods: We have developed a new method for segmentation of the head MRI that overcomes these limitations. As in the previous method, we have incorporated an atlas that is composed of an MRI image and tissue probability maps (TPM). The TPM assign each voxel in the atlas MRI with a value in the range of 0-1. This value reflects the probability that the specific tissue resides in that voxel. We have generated new customized TPMs for the cerebellum, brain stem and sinuses. Moreover, we have carefully revised the other TPMs to ensure best results. Last, we have incorporated these TPMs in a new atlas-based segmentation method. We have validated the method on 10 GBM patients T1w +gad head MRIs.

Results: The average Dice coefficient between gold-standard and algorithms' segmentation was 68.4% (SD=11.2%) that is similar to the value observed with the current method implemented in the treatment-planning software (67.8%; SD=10.6%). However, the new method presented here extends the current method with the cerebellum (average Dice = 84%) and brainstem (average Dice = 67%). Moreover, the sinuses are marked with air, and erroneous CSF segments in the sinuses were removed.

Conclusions: We have presented a novel method for the segmentation of the head that facilitates accurate infratentorial and supratentorial structures' segmentation. Moreover, the method eliminates erroneous CSF segmentation in the sinuses. These improvements pave the way for TTFields planning for infratentorial tumors.

Citation Format: Yana Glozman, Reuben R. Shamir, Zeev Bomzon. A method for infratentorial structures segmentation for tumor treating fields treatment planning [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 3071.

Abstract 1317: inovivo: a dedicated system for delivery of therapeutic level Tumor Treating Fields (TTFields) to mice

Purpose/objective: Tumor Treating Fields (TTFields) therapy is a noninvasive antineoplastic treatment modality that is FDA approved for treatment of glioblastoma and malignant pleural mesothelioma. TTFields are delivered to the patient continuously by 2 pairs of transducer arrays attached to the skin. In vivo TTFields experiments were so far limited due to the lack of a dedicated animal delivery system. Two main challenges are associated with TTFields application to the animal: 1) there is an absolute requirement for adequate and continuous contact between the electrodes and the animal skin throughout treatment; and 2) the wires connected to the electrodes require individual housing and limit animal movement, thus imposing stress. This work aimed to develop an in vivo system for continuous TTFields delivery to mice bearing cancer tumors in the torso or flank while addressing these challenges. Materials/Methods: To tackle the challenges and develop a viable in vivo system, several solutions were tested: 1) various electrode layouts; 2) a variety of adhesive materials; and 3) devices for preventing wire entanglement. Results: The final design of the transducer array electrodes included 2 adhesive layers, an inner layer for improved adherence, and an outer layer for securing the electrodes to the skin. Conductors were compacted into a single printed circuit cable connected to a novel electric swivel machine, that prevented cable coiling by sensing and rotating according to animal movement. These improvements resulted in fewer electrode entanglements and replacements, and thus in higher compliance (continuity) and less need for animal handling. To further reduce the impact of stress factors on the mice, a new cage was developed, that allows for 2 mice to be housed separately while still maintaining an interaction with one another. Indeed, animals treated with the inovivo system for 1-week displayed lower weight loss than animals treated with the previous non-dedicated system, indicative of reduced stress. Simulation were performed to ensure electric fields were indeed generated at the desired locations, showing above threshold TTFields intensities around the tumor for the flank subcutaneous model. For the torso orthotopic model, TTFields were shown to generate effective electric fields in the lung, liver, and pancreas, suggesting tumors in these organs may be treated effectively using the inovivo system. Conclusion: The new inovivo system provides means for continuous, 2 directions TTFields delivery to tumors in the torso or flank while minimizing stress on the mice. The inovivo system thus provides a tool for conducting TTFields experiments in mice, facilitating further in vivo studies for gaining additional mechanistical insight. The development of mouse head arrays to allow further research of the effect TTFields on glioblastoma, an application of widespread interest, is currently underway.

Citation Format: Shiri Davidi, Roni Blatt, Mijal Munster, Anna Shteingauz, Yaara Porat, Adel Zeidan, Tal Marciano, Zeev Bomzon, Moshe Giladi, Uri Weinberg, Yoram Palti. inovivo: a dedicated system for delivery of therapeutic level Tumor Treating Fields (TTFields) to mice [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 1317.

In Vivo Safety of Tumor Treating Fields (TTFields) Applied to the Torso

Background

Tumor Treating Fields (TTFields) therapy is a non-invasive, loco-regional, anti-mitotic treatment modality that targets rapidly dividing cancerous cells, utilizing low intensity, alternating electric fields at cancer-cell-type specific frequencies. TTFields therapy is approved for the treatment of newly diagnosed and recurrent glioblastoma (GBM) in the US, Europe, Israel, Japan, and China. The favorable safety profile of TTFields in patients with GBM is partially attributed to the low rate of mitotic events in normal, quiescent brain cells. However, specific safety evaluations are warranted at locations with known high rates of cellular proliferation, such as the torso, which is a primary site of several of the most aggressive malignant tumors.

Methods

The safety of delivering TTFields to the torso of healthy rats at 150 or 200 kHz, which were previously identified as optimal frequencies for treating multiple torso cancers, was investigated. Throughout 2 weeks of TTFields application, animals underwent daily clinical examinations, and at treatment cessation blood samples and internal organs were examined. Computer simulations were performed to verify that the targeted internal organs of the torso were receiving TTFields at therapeutic intensities (≥ 1 V/cm root mean square, RMS).

Results

No treatment-related mortality was observed. Furthermore, no significant differences were observed between the TTFields-treated and control animals for all examined safety parameters: activity level, food and water intake, stools, motor neurological status, respiration, weight, complete blood count, blood biochemistry, and pathological findings of internal organs. TTFields intensities of 1 to 2.5 V/cm RMS were confirmed for internal organs within the target region.

Conclusions

This research demonstrates the safety of therapeutic level TTFields at frequencies of 150 and 200 kHz when applied as monotherapy to the torso of healthy rats.

Abstract PO-076: First steps towards a framework for Tumor Treating Fields dosimetry and treatment planning

Introduction: In 2019 a study establishing the connection between Tumor Treating Fields (dose) at the tumor bed and overall survival in newly diagnosed Glioblastoma patients was published [1]. The investigators created realistic computational models of 340 newly diagnosed Glioblastoma patients that received TTFields as part of the EF-14 clinical trial. Computational models of the patients were created, and finite element simulations used to evaluate TTFields dose at the tumor bed comprising the region of enhancing tumor as well as a 3mm boundary zone surrounding the tumor and resection cavity. Their analysis showed a connection between the average field power of the field delivered to this region and patient survival. This study provide rationale for TTFields treatment planning, in which computational models are used in order to determine how best to place TTFields transducer arrays on the body to enhance patient outcome. Here we discuss a workflow we have developed for performing TTFields treatment planning a practical manner. Methods: TTFields treatment planning involves several steps including: (a) importing of patient imaging data into the treatment planning system (b) creation of an accurate patient model (c) defining a target region-the region in which TTFields dose should be optimized (d) optimization phase in which numerical simulations and optimization algorithms are used to optimal placement of the arrays (referred to as a treatment plan) (d) evaluation of the treatment plan and sign off. Here we present a practical framework for this procedure. Results: To framework utilizes a combination of T1c MRI and CT images in order to create a patient model. The images are first fused and the patient model created by semi-automatic segmentation of the images. Electric properties are assigned to each tissue type according to empirical values reported in the literature. The user selects the target region in which to optimize TTFields dose. Standard practice is to define a target region consistent with that used by Ballo et. al. [1]. Model quality is evaluated by the user, and if sufficient, optimization is initiated. During optimization, an iterative algorithm incorporating finite element simulations is used to identify optimal array layouts. The algorithm attempts to identify a layout that maximizes TTFields dose in the target region. Finally, the user is presented with a small set of (up to 5) layouts including the optimal layout. A set of color maps and dose Volume Histograms enable the user to select a treatment plan for the patient. Summary: We have developed a practical framework for TTFields treatment planning that in the future may be integrated into standard clinical practice when treating patients with TTFields.

Citation Format: Zeev Bomzon, Noa Urman, Shay Levi, Ariel Naveh. First steps towards a framework for Tumor Treating Fields dosimetry and treatment planning [abstract]. In: Proceedings of the AACR Virtual Special Conference on Radiation Science and Medicine; 2021 Mar 2-3. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(8_Suppl):Abstract nr PO-076.

TAMI-04. TUMOR TREATING FIELDS (TTFIELDS) HINDER GLIOMA CELL MOTILITY THROUGH REGULATION OF MICROTUBULE AND ACTIN DYNAMICS

The ability of glioma cells to invade adjacent brain tissue remains a major obstacle to therapeutic disease management. Therefore, the development of novel treatment modalities that disrupt glioma cell motility could facilitate greater disease control. Tumor Treating Fields (TTFields), encompassing alternating electric fields within the intermediate frequency range, is an anticancer treatment delivered to the tumor region through transducer arrays placed non-invasively on the skin. This novel loco-regional treatment has demonstrated efficacy and safety and is FDA-approved in patients with glioblastoma and malignant pleural mesothelioma. TTFields are currently being investigated in other solid tumors in ongoing trials, including the phase 3 METIS trial (brain metastases from NSCLC; NCT02831959). Although established as an anti-mitotic treatment, the anti-metastatic potential of TTFields and its effects on cytoskeleton rapid dynamics during cellular motility warrant further investigation. Previous studies have demonstrated that TTFields inhibits metastatic properties of cancer cells. Identification of a unifying mechanism connecting the versatile TTFields-induced molecular responses is required to optimize the therapeutic potential of TTFields. In this study, confocal microscopy, computational tools, and biochemical analyses were utilized to show that TTFields disrupt glioma cellular polarity by interfering with microtubule assembly and directionality. Under TTFields application, changes in microtubule organization resulted in activation of GEF-H1, which led to an increase in active RhoA levels and consequent focal adhesion formation with actin cytoskeleton architectural changes. Furthermore, the optimal TTFields frequency for inhibition of invasion in glioma cells was 300 kHz, which differed from the optimal anti-mitotic frequency leading to glioma cell death of 200 kHz. The inhibitory effect of TTFields on migration was observed at fields intensities of 0.6 V/cm RMS (below the threshold of 1 V/cm RMS previously reported for cytotoxic effects). Together, these data identify discrete TTFields effects that disrupt processes crucial for glioma cell motility.

CBIO-09. INTRATUMORAL HETEROGENEITY OF DIELECTRIC PROPERTIES IN GLIOBLASTOMA

INTRODUCTION

Recently, tumor treating fields (TTFields) were established for the treatment of newly diagnosed glioblastoma (GBM). One of the most crucial parameters defining the treatment efficacy of TTFields is the electric field intensity, which depends on the dielectric properties of the tumor tissue. In this study we determined the dielectric properties of GBM by analyzing resected tissue following a fast acquisition protocol. To account for the intratumoral heterogeneity, different regions of the tumor were analyzed separately.

METHODS

A cohort of 38 patients with newly diagnosed GBM were analyzed. Tissue probes were acquired from the vital tumor area and perinecrotic compartment. The tissue was measured immediately to avoid artifacts. A fragment was dissected from each tissue sample and was placed into a cylindrical cell with a known diameter. The impedance was recorded at frequencies 20Hz-1MHz using a software specifically developed for this study, which controls the LCR meter. The measured impedance was translated into dielectric properties of the sample (conductivity and relative permittivity) based on the parallel plate model, the recorded complex impedance and the geometry of the samples. Each tissue probe was fixed, and stained with H&E to visualize cellularity, luxol fast blue to analyze the myelinated fiber content and against factor VIII related antigen to assess tumor vascularity.

RESULTS

We found significant differences between the conductivity and permittivity of tissue samples from each individual tumor (mean conductivity [S/m]: 0.302; range: 0.607 – 0.100; mean permittivity [Farad/m]: 3519.8; range: 11182.5 – 135.7). Consistently, the perinecrotic areas displayed lower conductivity values compared to the solid tumor compartments. Histological analysis revealed significantly higher cellularity and lower myelinated fiber content in tissue samples with high conductivity and permittivity.

CONCLUSION

The dielectric properties of GBM show a high intratumoral heterogeneity which correlate to the extent of cellularity and myelin fiber content within the tissue.

NIMG-58. THE EFFECT OF CELL CONFLUENCE ON THE DISTRIBUTION OF TUMOR TREATING FIELDS

BACKGROUND

Tumor Treating Fields (TTFields) are known to exert anti-mitotic effects on cells. Numerical simulations investigating the electric field distribution within isolated cells have been reported. These studies have shown that during metaphase a uniform electric field forms within the rounded cells. This field is thought to disrupt spindle formation through alignment of the tubulin dimers with the field. Simulations also show that during cytokinesis, a non-uniform electric field forms at the furrow, leading to strong dielectrophoretic forces that further disrupt cell division. Cells in the tumor are densely packed. We used numerical simulations to investigate how the clustering of cells influences TTFields distribution.

METHODS

COMSOL was used to numerically simulate delivery of TTFields to clusters of round cells placed in a hexagonal arrangement. The influence of the distance between the cells on field distribution was investigated. The effect of pores in the cell membrane on field distribution was also investigated.

RESULTS

Placing round cells in clusters resulted in regions of highly non-uniform fields within the cells. Strong gradients in the electric field were also observed around pores placed in the membrane. Non-uniformity and gradients in the field could result in strong dielectrophoretic forces capable of disrupting key cellular structures such as the cytoskeleton and mitotic spindle, as well as cell membrane integrity.

CONCLUSIONS

The placement of cells in close proximity to one another creates gradients in the electric field, which could be associated with very strong dielectrophoretic forces that enhance the effects of TTFields on cells. Strong dielectrophoretic forces were also observed around the membrane pores. Previous studies have reported that TTFields increases membrane permeability [Chang et al Cell Death Discovery. 2018]. The strong dielectrophoretic forces in the membrane may provide a physical mechanism by which TTFields enhance membrane permeability.

RBIO-04. NEW THERAPEUTIC DELIVERY METHODS FOR TUMOR-TREATING FIELDS FOR HIGHER EFFICACY

Tumor-treating fields (TTFields) are the fourth modality of glioblastoma (GBM) treatment and in conjunction with chemotherapy can increase overall survival of GBM patients up to 60 months. However, in vitro and in animal models TTFields show 100% efficacy on a variety of tumor cell types including GBM cells when field strength is 4 V/cm, versus ~2 V/cm that is the clinical delivery target, TTFields are delivered transcranially. TTFields finite element modeling studies, supported by similar transcranial electric stimulation studies, show that the principal obstacle to delivering 4 V/cm is the electrically resistive skull. Our modeling shows the biophysics is more complicated than these findings. For instance, electrically-conductive cerebrospinal fluid regions surrounding the grey matter and in ventricles shunt electric current from anode to cathode, hindering delivery of the current required to produce 4 V/cm at the tumor/peritumor target. Thus, we consider two new delivery methods for TTFields. First, the transcranial array can be made more focal and directional, following modeling and development of electrode arrays used in spinal cord and deep brain stimulation. Our finite element modeling shows that similarly-designed TTFields electrode arrays can deliver field strength focally to a tumor target approaching 4 V/cm. Second, pre- or post-resection, TTFields can be delivered via electrode arrays surgically placed in the tumor or tumor resection cavity (intra-tumoral delivery), circumventing the resistive skull and CSF shunting effects. Such intra-tumoral arrays can deliver 4 V/cm to the tumor/peritumor region, opening up the potential to replicate clinically the 100% efficacy of TTFields in vitro and in animal models. Thus, new TTFields delivery may lead to unlimited survival of GMB patients via a side-effect free treatment modality.

NIMG-65. STUDY OF LOCAL PERTURBATION IN COMPUTATIONAL MODELLING ON TUMOR TREATING FIELDS (TTFIELDS) THERAPY

INTRODUCTION

Tumor Treating Fields (TTFields) are an approved therapy for glioblastoma (GBM). A recent study combining post-hoc analysis of clinical trial data and extensive computational modelling demonstrated that TTFields dose at the tumor has a direct impact on patient survival (Ballo MT, et al. Int J Radiat Oncol Biol Phys, 2019). Hence, there is rationale for developing TTFields treatment planning tools that rely on numerical simulations and patient-specific computational models. To assist in the development of such tools is it important to understand how inaccuracies in the computational models influence the estimation of the TTFields dose delivered to the tumor bed. Here we analyze the effect of local perturbations in patient-specific head models on TTFields dose at the tumor bed.

METHODS

Finite element models of human heads with tumor were created. To create defects in the models, conductive spheres with varying conductivities and radii were placed into the model’s brains at different distances from the tumor. Virtual transducer arrays were placed on the models, and delivery of TTFields numerically simulated. The error in the electric field induced by the defects as a function of defect conductivity, radius, and distance to tumor was investigated.

RESULTS

Simulations showed that when a defect of radius R is placed at a distance, d >7R, the error is below 1% regardless of the defect conductivity. Further the defects induced errors in the electric field that were below 1% when σrR/d < 0.16, where σrR/d < 0.16, where σr = (σsphere – σsurrounding)/(σsphere + σsurrounding).σsurroundings is the average conductivity around the sphere and σsphere is the conductivity of the sphere.

CONCLUSIONS

This study demonstrates the limited impact of local perturbations in the model on the calculated field distribution. These results could be used as guidelines on required model accuracy for TTFields treatment planning.

Abstract 472: What electric field strength is necessary for maximum tumor-treating fields efficacy

Why Is 4 V/cm Electric Field Strength Necessary for Maximum Tumor-Treating Fields Efficacy? INTRODUCTION: Tumor Treating Fields (TTFields) extend overall survival in glioblastoma patients. Threshold field strength for TTFields is typically cited as 2V/cm. Yet plots of cell death vs. field strength in vitro indicate a power law relationship asymptotic to 4 V/cm for 100% efficacy. The mechanisms of action (MoA) of TTFields underlying this relationship are currently under investigation. TTFields are suspected of disrupting critical mitotic processes performed by large polar cellular molecules such as microtubules (MT). METHODS: Using the power law relationship for the energy (intensity, I) carried by an electric field and its amplitude (Eq. 1), we calculated energy vs. TTFields amplitude. (1) RESULTS: We suggest Eq. 1 is a likely source of the relationship between TTFields' amplitude and cell death proportion that has been demonstrated empirically. Further, Table 1 shows energy transmitted per unit MT surface area per TTFields cycle, specifically, the energy per MT dimer band around an MT helix per TTFields period of 5 µs.

TTFields Field Strength (V/cm)Radiant Flux Density (W/nm2)Energy/(Cycle-Area) (J/nm2)Energy/MT Dimer/Cycle (J)Energy/MT Dimer Band/Cycle (J)113.3 × 10−186.63 × 10−234.01 × 10−224.17 × 10−20253.1 × 10−1826.6 × 10−231.60 × 10−211.67 × 10−193119 × 10−1859.7 × 10−233.61 × 10−213.75 × 10−194212 × 10−18106 × 10−236.42 × 10−216.67 × 10−195332 × 10−18166 × 10−231.00 × 10−201.04 × 10−186478 × 10−18239 × 10−231.44 × 10−201.50 × 10−18

A metric for disruption of cellular processes is 1-2 orders of magnitude greater than the thermal background energy in the cell, kT = 4.3 × 10−21 J. Table 1 shows that the energy absorbed by a MT dimer is insufficient to disrupt cellular processes at 2 V/cm, while due to energy increasing with square of field strength, at 5 V/cm it may be disruptive. For an entire dimer band around the MT helix, 2 V/cm appears to be sufficient, while 4 V/cm seems amply strong to disrupt MT function, in accordance with in vitro experiments. CONCLUSION: Assuming the power law of energy vs. electric field strength and the resulting energy absorbed by microtubule components, we found a correlation between typical TTFields' amplitudes and estimated disruption thresholds of MT function. TTFields MoA. 4 V/cm, not 2V/cm, should be the target field strength to realize the full efficacy of TTFields.

Citation Format: Kristen W. Carlson, Jack A. Tuszynski, Socrates Dokos, Nirmal Paudel, Zeev Bomzon. What electric field strength is necessary for maximum tumor-treating fields efficacy [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 472.

Compatibility of topical agents with tumor treating fields (TTFields) for treatment of associated skin events in glioblastoma (GBM)

e24126

Background: TTFields are low intensity, intermediate frequency, alternating electric fields applied continuously using 2 pairs of skin-affixed transducer arrays. TTFields are FDA-approved in GBM and mesothelioma. The most common TTFields-related adverse event (AE) is mild-to-moderate dermatitis (beneath arrays), via long-term irritant exposure and local hyperhidrosis and occlusion exacerbation. Skin reaction mitigation strategies may improve quality-of-life (QoL) and ensure usage, as maximal survival benefits have been correlated with duration of use. Not all skin care products are TTFields compatible and may increase electrical impedance and lead to beneath array temperature increases. The aim of this in vivo study was to investigate the effects of 62 commercially available skin care products on electrical impedance during TTFields treatment. Methods: TTFields (200 kHz; optimal GBM frequency) were applied to rats using transducer arrays made of the same ceramic disks and hydrogels used in patients with GBM. To test electrical impedance effects, skin care products were applied to the skin immediately before array placement. The change in impedance relative to naïve skin was measured using the Optune device. Sixty-two commercially available products from 8 groups (adhesive removers, antibiotics, antiperspirants, antiseptics, cleansers, moisturizers, skin barriers, and topical corticosteroids) were evaluated. Results: Most lotions, soaps, foams, and solutions had minimal effect on electrical impedance, while petrolatum-based ointments significantly increased electrical impedance. Conclusions: TTFields compatible skin care products that did not affect electrical impedance were identified from each of the 8 groups and could be considered for further evaluation. All petrolatum-based ointments that were tested led to an increase in electrical impedance and are thus not recommended. Local application of TTFields compatible skin care products should be prospectively investigated in the clinical setting for their potential role in minimizing TTFields–related skin AEs. The randomized, double-blind PROTECT (PROphylactic skin Toxicity thErapy with Clindamycin and triamcinolone in GBM patients Treated with TTFields) study, should help establish which products best reduce skin AEs in patients with GBM and assess impact on QoL.

P11.37 Evaluating water content and electrical properties at 200 kHz in brain and GBM tumor tissue of three TTFields patients with conventional imaging

BACKGROUND

Electrical properties (EPs) of brain tissue, specifically brain tumors, crucially determine the field distribution of Tumor Treating Fields (TTFields), an anti-mitotic treatment approved for glioblastoma multiforme (GBM). Due to the correlation of TTFields efficacy and field intensity at the tumor region, the knowledge of EPs in each patient is of great importance for patient-specific planning of treatment. Water content electrical properties tomography (wEPT) is a non-invasive imaging technique using water content (WC) maps obtained from rapidly acquired and processed conventional sequences to estimate the EPs of brain tissue at 128 MHz. The WC maps of this approach are constructed from two spin echo sequences similar to a T1 and a PD image. Following previous studies in rat tumor models demonstrating promising wEPT mapping of EPs in the brain at 200, this study examines the feasibility of this approach in human GBM patients.

MATERIAL AND METHODS

For three patients of the EF-14 trial population, we divided T1 and PD images pixel-by-pixel to obtain the image ratio. Using a transfer function, WC maps were generated and maps of the electrical conductivity σ and the relative permittivity ε r at 200 kHz were calculated with two different equations.

RESULTS

The median value of estimated WC remains similar in healthy brain tissues among all patients, ~73.5% in the white matter, ~82% in the gray matter. The median values of wEPT-estimated σ at 200 kHz in the white matter is ~0.09 S/m and in the gray matter ~0.18 S/m, corresponding median values of ε r at 200 kHz are ~2100 and ~3000 in white and gray matter respectively. Contrary, in the tumor the spread between the median values of WC and EPs is much higher. Stating the most important findings, in the necrosis median WC are 90.3%, 92.3%, 85.2% in patients 1–3 respectively with corresponding median σ values of 0.494, 0.657, 0.25 S/m. In the enhancing tumor the spread of median WC is even higher (67.2%, 83.6%, 85.5%), yet lower spread but also very heterogeneous median σ values of 0.075 S/m, 0.208, 0.259 S/m are estimated with wEPT.

CONCLUSION

Our results demonstrate the adaption of wEPT for mapping of WC and EPs at 200 kHz in three human GBM patients. In contrast to the vastly irregular tumor tissue, our estimations in healthy brain tissue are similar between patients and in accordance with EPs experimentally measured during our animal experiments and consistent with reported values in the literature. Hence, wEPT is a promising, fast technique based on regular MRI that might help patient-specific treatment planning of TTFields therapy, although the mapping of tumor tissue needs further confirmation in a greater population and investigations of EPs of excised tumor tissue samples should be conducted.

P11.50 The dielectric properties of brain tumor tissue

BACKGROUND

Recently, tumor treating fields (TTFields) were established for the treatment of newly diagnosed GBM. One of the most crucial parameters defining the treatment efficacy of TTFields is the electric field intensity. The dielectric properties of the normal intracranial compartments are well established, allowing the prediction of the electric field distribution. In contrast, there is no data available about the dielectric properties of brain tumor tissue. In this study we determined the dielectric properties of brain tumors by analyzing resected tissue following a fast acquisition protocol. To account for the intratumoral heterogeneity, different regions of the tumor were analyzed separately.

MATERIAL AND METHODS

A cohort of 84 patients with tumors of different histology and malignancy grade have been recruited (meningioma: n=26; brain metastases n=18; low grade glioma n=6; glioblastoma n=34). Tissue probes were acquired whenever possible from the vital tumor area, and perinecrotic compartment identified intraoperatively using neuronavigation, intraoperative ultrasound and fluorescence guidance. After acquisition, the tissue was measured immediately to avoid artifacts induced by temperature change, differences of fluid composition as well as post resection ischemia. A fragment was dissected from each tissue sample and was placed into a cylindrical cell with a known diameter. Two parallel electrodes were placed on both sides of the sample and the thickness of the tissue was measured using a micrometer. The impedance was recorded at frequencies 20Hz-1MHz using a software specifically developed for this study, which controls the LCR meter (Keysight Technologies, Santa Rosa, USA). The measured impedance was translated into dielectric properties of the sample (conductivity and relative permittivity) based on the parallel plate model, the recorded complex impedance and the geometry of the samples. Each tissue probe was fixed, H&E stained and histologically analyzed.

RESULTS

We found significant differences between the conductivity of different types of tumors with meningiomas showing the lowest conductivity (mean conductivity [S/m]: 0.193; range: 0.327 - 0.113) and GBM tissue exhibiting the highest conductivity values (mean conductivity [S/m]: 0.402; range: 0.893 - 0.157). Consistently, the perinecrotic areas of tumors displayed lower conductivity values compared to the solid tumor compartments. Also, we found a significant intratumoral heterogeneity in tumors of one specific histological diagnosis.

CONCLUSION

The dielectric properties of intracranial tumors appear to be depending on histological class and malignancy grade and show significant intratumoral heterogeneity. These results may allow a more precise modelling of electric field intensity distribution within the tumor.

P14.69 Evaluation of electric field intensity delivered by Tumor-Treating Fields therapy to PET-defined metabolic volumes in recurrent glioblastomas

BACKGROUND

Tumor-Treating Fields (TTFields) therapy is a clinical treatment option for patients with newly-diagnosed and recurrent glioblastomas. Electric field intensities (EFIs) delivered to the tumor mass may affect treatment responses. In this study, we used the patients’ neuroimaging data to create realistic head models and evaluate: (i) the magnitude of EFIs delivered to the tumor mass; (ii) factors affecting the EFI values; and (iii) factors affecting treatment responses as assessed by amino acid PET.

MATERIAL AND METHODS

Fourteen recurrent glioblastomas in 9 patients were evaluated with α-[11C]-methyl-L-tryptophan (AMT)-PET before and up to 3 months after TTFields therapy (mean follow-up: 2.3 months). Individual MRI and CT scans were used to create patient-specific realistic head models and simulate TTFields delivery to the tumors. For each direction of treatment (antero-posterior, left-right), two 9-disk transducer arrays were simulated using disks placed according to the patients’ NovoTAL System™ based treatment plan. To generate TTFields, an alternating voltage difference (200V peak-to-peak, 200 kHz) was imposed on the outer surfaces of the disks. The simulations were performed using the Sim4Life V3.0 (ZMT-Zurich) quasi-electrostatic solver. The field intensities were normalized to simulate 2A peak-to-peak current supplied by the device. 3D EFI maps were created and fused with the pre- and post-TTFields PET images to measure EFIs delivered to the PET-defined metabolic tumor volume. Interval changes of static AMT uptake and kinetic PET variables were also evaluated.

RESULTS

The mean EFI delivered to the tumors varied between 1.34–2.43 V/cm (mean: 1.86 V/cm). Fronto-parietal tumors received higher mean EFI than temporal lobe tumors (p=0.05). Most tumors showed decreasing (n=9) or stable (n=4) AMT uptake on follow-up PET imaging after TTFields therapy. Higher EFIs delivered to the tumors (r=-0.56, p=0.04) and concomitant bevacizumab treatment (n=7, p=0.01) were associated with a greater PET response. On tracer kinetic analysis, the AMT uptake responses correlated with transport rate changes (p=0.04).

CONCLUSION

TTFields treatment of recurrent glioblastomas delivers variable EFIs to the metabolic tumor volume. Treatment responses on PET are driven by decreased amino acid transport rates, whose magnitude is associated with higher EFIs delivered to the tumor mass and also with concomitant antiangiogenic treatment in those with combined therapy. (The cost of the PET scans was supported by a grant from NovoCure Ltd., Haifa, Israel)

P11.18 Tumor treating fields (TTFields) treatment of spinal cord metastases

BACKGROUND

Tumor Treating Fields (TTFields) is an anti-mitotic cancer treatment approved for the treatment of Glioblastoma multiforme (GBM) and is currently also investigated in a phase III trial in 1–10 brain metastases from non-small cell lung cancer (METIS). Apart from spread to the brain, some cancer types, such as breast cancer, lung cancer, and melanoma, may lead to metastatic spread to the spinal cord. Previous studies have shown that reported transducer array layouts for the treatment of abdominal/pelvic tumors (e.g. pancreatic cancer), with one pair of arrays positioned on the anterior and posterior of the patient, and the second pair of arrays placed on each side of the thorax, yield therapeutically insufficient field intensities of <1 V/cm in the spinal cord. This finding probably results from the anatomical structure of the spine, consisting of the cerebrospinal fluid as a highly conductive layer, encased by a resistive bone structure that shunts the current delivered across the body by the arrays away from the spine. This simulation-based study aimed at resolving this challenge by identifying novel array layouts on the body that effectively deliver TTFields to the spine.

MATERIAL AND METHODS

For the simulations of the TTFields delivery to the spine, a human male 34 years old realistic computational model (DUKE v3.1 by ITI’S, Zurich) and the ZMT’s Sim4Life v4.0 electro-quasi-static solver was utilized. TTFields were simulated by imposing an alternating current with a current density of 200 mA/disk and a frequency of 150 kHz on the outer surfaces of the disks of each pair of arrays.

RESULTS

For one of the tested array layouts, a high electric field was shown to be induced within the spinal cord and surrounding CSF: Our calculations of mean field intensity within the spine and nerves from vertebrae T8-T9 at the top to L3-L4 at the bottom added up to 1.77 V/cm. This layout consisted of the placement of a pair of arrays on the back of the patient, with one array positioned above the section in the spine to which treatment would be delivered, and the other array positioned below the target section. Notably, the resulting electric field is directed along the spine in this setting (ie, vertically).

CONCLUSION

Our results demonstrate that treatment of the whole spinal cord and nerves in a single direction can be achieved by placing a pair of transducer arrays on the patient’s back: one array on the neck, and one at the bottom of the spine. For the development of an active treatment in the perpendicular direction, further studies need to be conducted.

P11.25 Assessing electrical properties of cells as predictive marker for patient-specific TTFields response and optimal frequency

BACKGROUND

Tumor treating fields (TTFields) are currently approved for the treatment of glioblastoma multiforme (GBM, using 200 kHz), and being tested in other tumor types such as non-small cell lung cancer and brain metastases occurring in this indication (LUNAR and METIS trials, using 150 kHz). Response to TTFields in cancer cells was empirically shown to be frequency-dependent specific for cell type; however, there are no markers available predicting optimal frequency or response in different cancer types or individual patients to date. There is evidence indicating electrical properties determine the optimal anti-mitotic frequency. This study analyzed the correlation of electrical properties of cells with their optimal TTFields frequency and sensitivity using the 3DEP reader (LabTech) to determine the electrical properties with the help of Dielectrophoresis (DEP) force. With this technique, cell movements within electric fields of different frequencies can by analyzed based on the physical effect of DEP, exercising a force on polarizable particles inside a non-homogeneous electric field.

MATERIAL AND METHODS

We used the 3DEP reader to obtain baseline properties (permittivity and conductivity) of 17 different cell lines of several tumor types. The resulting curves were analyzed by a 2-way ANOVA. Additionally, we determined the optimal frequency for maximum cytotoxic effect for each cell line using the inovitroTM system and eventually compared with the detected electrical properties.

RESULTS

We found cell lines with an optimal TTFields frequency of 150 kHz (corresponding to cells with a membrane capacitance in the lower range of the observed 3DEP curves, n=9) to possess significantly different (p<0.001) electrical properties from cells with an optimal TTFields frequency of 200 kHz (n=8). According to the curve differences in the lower frequency range, the measure of membrane capacitance served as a good predictor for TTFields response.

CONCLUSION

This study showed a correlation of cell membrane capacitance and optimal TTFields frequency and response. Our results provide a substantial rationale for further studies investigating the predictive potential of electrical properties of tumor cells as a measure for the optimal frequency and sensitivity to TTFields in individual patients.

Abstract 4031: The dielectric properties of brain tumor tissue

Introduction: Recently, tumor treating fields (TTFields) have been established as for the treatment of newly diagnosed GBM1. One of the most crucial parameters defining the treatment efficacy of TTFields is the electric field intensity. The electric properties of the normal intracranial compartments are well established, allowing the prediction of the electric field distribution. In contrast, there is no data available about the electric properties of tumor tissue. In this study we determined the dieclectric properties of brain tumors by analyzing resected tissue following a fast acquisition protocol. To account for the intratumoral heterogeneity, different regions of the tumor were sampled and analyzed separately.

Methods: A cohort of thirty patients with tumors of different histology and malignancy grade have been recruited (meningioma: n=11; brain metastases n=7; low grade glioma n=6; glioblastoma n=6). Tissue probes were acquired whenever possible from the vital tumor area, and perinecrotic compartment identified intraoperatively using neuronavigation, intraoperative ultrasound and fluorescence guidance. From each region, five tissue probes were sampled. After acquisition, the tissue was measured immediately to avoid artifacts induced by temperature change, differences of fluid composition as well as post resection ischemia. A fragment was dissected from each tissue sample and was placed into a cylindrical cell with a known diameter. Two parallel electrodes were placed on both sides of the sample and the thickness of the tissue was measured using a micrometer. The impedance was recorded at frequencies 20Hz-1MHz using a software specifically developed for this study, which controls the LCR meter (Keysight Technologies, Santa Rosa, USA). The measured impedance was translated into dielectric properties of the sample (conductivity and relative permittivity) based on the parallel plate model, the recorded complex impedance and the geometry of the samples. Each tissue probe was fixed, H&E stained and histologically analyzed for tumor cell count and specific tissue features.

Results: After receiving a positive ethics votum, the first 30 patients were recruited. As a reference, grey and white matter tissue samples from mouse brains were used. We found significant differences between the tumor entities with meningiomas showing the lowest and GBM tissue the highest conductivity values. Consistently, the perinecrotic areas displayed lower conductivity values compared to the solid tumor compartment. Also, we found a significant heterogeneity in the dielectric properties within one tumor. Conclusion: The dielectric properties of intracranial tumors appear to be depending on histological class and malignancy grade and show significant intratumoral heterogeneity. These results may allow a more precise modelling of electric field intensity distribution within the Tumor.

Citation Format: Martin A. Proescholdt, Amer Haj, Christian Doenitz, Alexander Brawanski, Zeev Bomzon, Hadas Sara Hershkovich. The dielectric properties of brain tumor tissue [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4031.

Abstract 3725: Numerical simulation of tumor treating fields effects on cell structures: Mechanism and signaling pathway candidates

Tumor Treating Fields (TTFields) have become a fourth modality for cancer treatment. Mild electric fields (~1-4 V/cm) produce few side effects and significantly extend overall survival of glioblastoma patients, and TTFields are in clinical trials for a variety of tumor cell types. Our goal is to uncover TTFields’ mechanism and cell signaling pathways by numerically modeling their effects on sub-cellular structures, such as microtubules (MTs) and their interactions with motor proteins. METHODS: We have built finite element models in COMSOL Multiphysics (tm) of the MT and its micro-environment to test hypotheses on TTFields’ mechanism of action by predicting effects on sub-cellular structures. RESULTS: One model prediction is that current density induced in the MT counter-ion layer by TTFields essentially shunts electric current within them. The strongest current flows through the counter-ion layer surrounding the MT’s C-termini and energy density in this layer likely exceeds the level to disrupt motor protein ‘walk’ along the MT. The energy density is predicted at 10-20 Joules when both the field and the MTs are aligned with the cell axis. A second mechanism examined by our model is disruption of the ‘foot’ of kinesin, released from its C-terminus contact by ATP (10-19 Joules). The final phase of the walk is driven by thermal buffeting of the forward foot randomly positioning it near enough to the C-terminus for electrostatic forces to bind it. A stall force ~10-19 - 10-16 N from TTFields would prevent diffusion and disrupt the kinesin walk. A recent clinical study segregating patient cohorts treated vs. not treated with dexamethasone found overall survival indefinitely for the non-dexamethosone cohort, leading us to hypothesize that TTFields activate the intrinsic Bcl2-mediated apoptotic signaling pathway. Future modeling will seek to tie disruption of motor protein transport along MTs to activating intrinsic apoptosis, e.g. via failure to silence the G2 cell cycle checkpoint. CONCLUSION: Our modeling predicts that TTFields in cytosol induce electric currents along MTs that are strong enough to disrupt key cellular functions such as the kinesin walk and C-termini transitions, both of which are crucial for motor protein transport. Hence, TTFields disrupt the most delicate mechanisms involved in the carefully-orchestrated succession of steps in mitosis.

Citation Format: Kristen W. Carlson, Nirmal Paudel, Jack A. Tuszynski, Zeev Bomzon. Numerical simulation of tumor treating fields effects on cell structures: Mechanism and signaling pathway candidates [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3725.

Abstract 1413: Using conventional imaging to predict water content and electrical properties at 200 kHz in brain and GBM tumor tissues: a feasibility study in three TTFields patients

Introduction

Tumor Treating Fields (TTFields) are approved for the treatment of Glioblastoma (GBM). The efficacy of TTFields increases with field intensity at the tumor which depends on its highly heterogeneous electrical properties (EPs) distribution. Thus, an imaging technique, preferably with conventional sequences, rapid acquisition and fast processing, that is able to assess the EPs non-invasively might be of high interest for patient-specific TTFields treatment planning.

An approach termed water content electrical properties tomography (wEPT), estimates the EPs of brain tissue at 128 MHz from water content (WC) maps which are created with two spin echo sequences resembling a T1w and a PD image. Recently we performed experiments in tumor-bearing rats suggesting that wEPT could be adapted to map EPs in the brain at 200 kHz. Here we tested the feasibility of applying wEPT to map EPs at 200 kHz in GBM patients.

Methods

Analysis was performed for three patients that participated in the EF-14 trial. The image ratio was calculated as pixel-by-pixel division of T1 and PD images. The WC maps were estimated with a transfer function and two separate equations were used to calculate the maps of the electrical conductivity σ and the relative permittivity ϵr at 200 kHz. The table summarizes the median values in 5 tissues of normalized T1 and PD signals and the wEPT-estimations of WC, σ, and ϵr at 200 kHz.

Results

median valuesT1normPDnormWC[%]σ[S/m]ϵr[-]white matterPatient 10,230,5876%0,112575Patient 20,180,3276%0,112462Patient 30,240,3574%0,102191gray matterPatient 10,200,6083%0,222923Patient 20,160,3381%0,172799Patient 30,220,3881%0,172919necrosisPatient 10,230,8689%0,472495Patient 20,120,3298%1,431381Patient 30,240,4484%0,242988enhancing tumorPatient 10,370,8371%0,091546Patient 20,170,3581%0,172961Patient 30,230,4285%0,262932non-enhancing tumorPatient 10,230,7385%0,262979Patient 20,180,3678%0,142769

Conclusions

We adapted the wEPT approach to map WC and EPs at 200 kHz in three GBM patients. In healthy brain tissues the estimations are consistent with literature and also among patients, contrary to the highly variable tumor. This fast approach only needing conventional MRI holds some promise for patient-specific TTFields treatment planning. Yet the mapping of tumor tissues needs to be validated further, possibly including EP measurements of excised tumor samples and generally a higher number of patients analyzed.

Citation Format: Cornelia Wenger, Hadas Sara Hershkovich, Catherine Tempel Brami, Moshe Giladi, Zeev Bomzon. Using conventional imaging to predict water content and electrical properties at 200 kHz in brain and GBM tumor tissues: a feasibility study in three TTFields patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1413.

Abstract 688: Safety and efficacy of TTFields delivery to the lungs: A computational study

Objective: This study investigates the efficacy and safety of Tumor Treating Fields (TTFields) delivery to the lungs utilizing computational simulations.

Introduction: Recent results of a phase 2 clinical trial STELLAR (NCT02397928) demonstrated a significant extension in median overall survival among patients with mesothelioma treated with TTFields plus standard of care chemotherapy compared to historical control data of patients who received standard of care chemotherapy alone. A phase 3 clinical trial (NCT02973789) is currently investigating the efficacy of TTFields therapy for the treatment of Non-Small-Cell lung cancer (NSCLC). The efficacy of TTFields therapy in disrupting tumor cells depends on the frequency of the field (150kHz optimal frequency for NSCLC and mesothelioma) and the field intensity. The higher the field intensity, the larger the therapeutic effect, with a therapeutic threshold of 0.7 V/cm above which TTFields begin to exert a significant anti-mitotic effect on cells.

Delivery of an electric field to the body unavoidably leads to deposition of heat in the tissue through Joule heating. The rate at which electrical energy is absorbed and converted to heat, depends on field’s frequency and tissue’s properties. Since Joule heating may cause thermal damage, it is important to ensure that TTFields intensity in the body is below the thermal damage threshold. In this study we performed numerical simulations to evaluate the thermal safety and the efficacy of the NovoTTF-100L when delivering Tumor Treating Fields to the torso. Thermal safety threshold levels were determined by current density and specific absorption rate (SAR) and TTFields therapeutic threshold was determined by field intensity.

Methods: We investigated field intensity, current density and SAR values that developed within 3 different realistic human computational models in which a virtual representation of the NovoTTF-100L (150kHz) was used to deliver TTFields to the lungs. The models used in this study include: a female model, male model and an obese male model (Virtual Population, IT’IS foundation) with a range of BMI values from normal to obese. Numerical simulations were performed using Sim4life (v3.0, ZMT Zurich).

Results: The simulations show that for all models, the NovoTTF-100L delivers therapeutic intensities of over 0.7 V/cm RMS to over at least 76% of the lungs. Current density within the models is well below the safety threshold of 100 mA/cm^2. SAR values within the internal organs are below the levels at which thermal damage occurs. In the superficial body layers, higher SAR values are observed. However, the NovoTTF-100L incorporates temperature control that prevents the skin from heating to levels at which thermal damage can occur.

Conclusions: Results of this study support the observations that the NovoTTF-100L delivers TTFields to the lungs at therapeutic levels and that the device is safe for use in the entire patient population.

Citation Format: Hadas Sara Hershkovich, Noa Urman, Ariel Naveh, Zeev Bomzon. Safety and efficacy of TTFields delivery to the lungs: A computational study [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 688.

Abstract CT176: Phase III PANOVA study of tumor treating fields (150 kHz) in combination with nab-paclitaxel and gemcitabine for front-line treatment of locally-advanced pancreatic adenocarcinoma (LAPC)

Background: Tumor Treating Fields (TTFields) are a non-invasive, regional antimitotic treatment modality, which has been approved for the treatment of patients with glioblastoma by the FDA. TTFields predominantly act by disrupting the formation of the mitotic spindle during metaphase. TTFields were effective in multiple preclinical models of pancreatic cancer. The Phase II PANOVA study [NCT01971281] of TTFields in pancreatic cancer (metastatic and LAPC) combined with nab-paclitaxel and gemcitabine, demonstrated no TTFields-related serious adverse events; key TTFields-related AEs was manageable skin toxicity. TTFields plus gemcitabine and nab-paclitaxel showed median PFS was 12.7 months. The Phase III PANOVA-3 trial [NCT03377491] is designed to test the efficacy of adding TTFields to nab-paclitaxel and gemcitabine combination in LAPC.

Methods: Patients (N = 556) with unresectable, LAPC (per NCCN guidelines) will be enrolled in this prospective, randomized trial. Patients should have an ECOG score of 0-2 and no prior progression or treatment. Patients will be stratified based on their performance status and geographical region, and will be randomized 1:1 to TTFields plus nab-paclitaxel and gemcitabine or to nab-paclitaxel and gemcitabine alone. Chemotherapy will be administered at standard dose of nab-paclitaxel (125 mg/m2) and gemcitabine (1000 mg/m2 once weekly). TTFields (150kHz) will be delivered at least 18 hours/day until local disease progression per RECIST Criteria V1.1. Follow up will be performed q8w, including a CT scan of the chest and abdomen. Following local disease progression, patients will be followed monthly for survival. Overall survival will be the primary endpoint and progression-free survival, objective response rate, rate of resectability, quality of life and toxicity will all be secondary endpoints. Sample size was calculated using a log-rank test comparing time to event in patients treated with TTFields plus chemotherapy with control patients on chemotherapy alone. The PANOVA-3 study is designed to detect a hazard ratio 0.75 in overall survival. Type I error is set to 0.05 (two-sided) and power to 80%.

Citation Format: Uri Weinberg, Ori Farber, Moshe Giladi, Zeev Bomzon, Eilon Kirson. Phase III PANOVA study of tumor treating fields (150 kHz) in combination with nab-paclitaxel and gemcitabine for front-line treatment of locally-advanced pancreatic adenocarcinoma (LAPC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr CT176.

Abstract 2168: Testing the electrical properties of different cell lines using 3DEP reader and compare to TTFields response

Background: Tumor Treating Fields (TTFields) are an approved anti-neoplastic treatment modality delivered via application of low intensity, intermediate frequency, alternating electric fields. The electrical properties of cells (such as permittivity and conductivity) determine the optimal frequency of TTFields that incurs the highest reduction in cell counts. Currently, there are no predictive markers for determining TTFields response or the optimal frequency to be applied for individual patient. The goal of this study is to determine the correlation between electrical properties of cells and TTFields’s optimal frequency and sensitivity. The 3DEP reader (LabTech) determines the electrical properties of cells, including permittivity and conductivity, by using Dielectrophoresis (DEP) force. DEP is a physical effect that generates a force on polarizable particles experiencing a non-homogeneous electric field and can therefore be used as a technique to analyse the way cells move within electric fields at different frequencies. Methods: The baseline electrical properties (permittivity and conductivity) of 17 cell lines from different tumor types were determined using the 3DEP reader (LabTech). The curves were analyzed using 2-way ANOVA. The optimal TTFields frequency for each cell line was determined by testing the cytotoxic effect of TTFields at various frequencies using the inovitro system. The electrical properties of cells were compared with the optimal TTFields frequency and sensitivity of each cell line. Results: The results demonstrate significant differences (p&lt;0.001) between the lower frequency range of the 3DEP curves that corresponds to membrane capacitance of cells with TTFields optimal frequency of 150kHz (9 cell lines) and cells with TTFields optimal frequency of 200kHz (8 cell lines). Membrane capacitance was also a good predictor for TTFields sensitivity based on the differences in the curves in the low frequency range. Conclusions: The results presented in this study demonstrate that cell membrane capacitance correlates with TTFields optimal frequency and sensitivity. Based on the above, there is a strong rational to further explore the potential of measuring the electrical properties of cells as a predictive marker to help determine which patient will respond better to TTFields and the optimal TTFields frequency to be applied for each patient.

Citation Format: Moshe Giladi, Einav Zeevi, Karnit Gotlib, Cornelia Wenger, Ariel Naveh, Zeev Bomzon, Eilon D. Kirson, Uri Weinberg, Adrian Kinzel, Yoram Palti. Testing the electrical properties of different cell lines using 3DEP reader and compare to TTFields response [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2168.

A general approach to optimizing tumor treating fields therapy

e14668

Background: Tumor Treating Fields (TTFields) are alternating electric fields used to non-invasively treat cancer. TTFields are delivered via transducer arrays placed on the skin close to the tumor. Post-hoc analysis [1] has shown that delivering higher field power to the tumor and increasing usage (percent of time patient is actively treated) improve patient survival. Thus, optimizing the position of arrays to maximize TTFields power at the tumor could improve survival. At the same time, minimizing the array area to maximize patient comfort and consequently maximizing usage is also likely to improve survival. However, optimizing TTFields delivery is non-trivial since the field distribution is influenced by array positioning and geometry, the anatomy of the patient and the heterogeneous electric properties of different tissues. Here we present a general approach to optimizing Tumor Treating Fields using numerical simulations. Methods: Delivery of TTFields to the brains, lungs and abdomens of realistic computational models was investigated. The effect of the transducer array size and position on the field distribution within the phantoms was analyzed, and an approach for optimizing TTFields delivery developed. Results: Field power is generally highest in the region between the arrays, with larger arrays generally delivering higher field power. Anatomical features such as bones, the spine or a resection cavity significantly influence the field within this region. A general approach to optimizing TTFields delivery is: Maximize field power by using the largest arrays possible. To maximize patient comfort, array size are chose so that significant portions of the skin in the region of disease are not covered by the arrays. Place virtual arrays on a realistic computational model of the patient such that the tumor is located between them and simulate TTFields delivery to the patient. Apply an iterative algorithm to shift the arrays around their initial positions until field power in the tumor bed is maximized. Conclusions: We have developed a general approach to optimizing delivery of TTFields to the tumor. Effective TTFields treatment planning is expected to improve patient outcome. [1] Ballo et. al., submitted to RED Journal 2018.

Transducer array layout optimization for the treatment of pancreatic cancer using Tumor Treating Fields (TTFields) in the phase 3 PANOVA-3 trial

e15766

Background: TTFields is an antimitotic cancer treatment that utilizes low intensity (1-3 V/cm) alternating electric fields in the intermediate frequency (100-300 kHz), which are delivered in two orthogonal directions using 2 pairs of transducer arrays. Based on favorable results in a Phase 2 study in locally-advanced pancreatic cancer (LAPC), an ongoing Phase 3 PANOVA-3 trial [NCT03377491] is investigating the efficacy of adding TTFields to nab-paclitaxel and gemcitabine in LAPC. Preclinical studies show that the effect of TTFields is intensity-dependent with a therapeutic threshold of 1 V/cm. The field distribution within the body is known to changes with array placement. The current study was designed to develop clinical practice guidelines for optimizing the layout of arrays applied to patients participating in the TTFields arm of the PANOVA-3 study. Methods: Three realistic computerized models of a male, a female and an obese male were used to simulate delivery of TTFields to the abdomen. For each model, 6-8 different layouts utilizing combinations of arrays with either 13 or 20 disks per-array were tested. The arrays were placed over the upper 6 standard abdominopelvic regions, and field intensity distributions within these regions were evaluated. Results: In all simulations, the large arrays generated higher field intensities than the smaller arrays. However, models with lower BMI were almost covered entirely using large arrays, increasing the potential of skin toxicity from TTFields. However, these Low BMI models were able to receive TTFields at an anti-mitotic intensity with smaller arrays. The clinical guidelines were formulated based on the following principles: the target tumor region should be directly between the arrays, (b) the extent of disease as well as the anatomy of the patient determines the size of arrays determined using waist circumference measurements. Conclusions: A matrix for selecting between 8 individual array layouts was generated based on these principles. The clinical practice guidelines allow the optimization of TTFields delivery to individual patients treated in the PANOVA-3 Study.