Dosimetric Evaluation of a Novel Non-Metallic Dual Port Expander for Proton Therapy

PURPOSE/OBJECTIVE(S)

This study evaluates a novel non-metallic (plastic) dual port expander and clinically utilized single/dual port metallic expanders for intensity modulated proton therapy (IMPT) in order to quantify the dosimetric implications in clinical treatment plans.

MATERIALS/METHODS

Four previously treated breast cancer patients (3 left sided, 1 right sided) with tissue expanders were retrospectively planned for IMPT utilizing a treatment planning system. Utilizing the existing CT data sets, models of three Sientra tissue expanders (dual port plastic (DPP), dual port metal (DPM) and single port metal (SPM)) were inserted into the 3D dataset and aligned using the existing patient anatomy. Each patient was planned to a total dose of 50.4 Gy in 28 fractions using 3 incident proton fields and Monte Carlo calculation algorithm. A 5 mm expansion was placed around the metallic portion of the expander to define a beam avoidance region which no proton spots traversed or were placed to minimize dosimetric error. Robust optimization (5mm setup uncertainty, 3.5% range uncertainty) was employed to a total target volume which comprised the CTV chest wall, IMN, axilla and supraclavicular nodes. The subsequent treatment plans were evaluated for target and OAR coverage.

RESULTS

Due to the non-metallic construction of the DPP expander, the metallic structure avoidance region was significantly smaller at 4.4cc on average than either the SPM (45cc average) or DPM (63.5cc average). This allowed the IMPT plans to be significantly less modulated, delivering a more uniform dose to the implant volume (SPM 47.1Gy; DPM 45.2 Gy; DPP 50.9 Gy) and thus enabling these plans to be inherently more robust to setup uncertainties. The additional benefit of the reduced avoidance region size is that the hinge angle of the lateral proton beams can be reduced for DPP plans thus reducing dose to the heart and contralateral breast. Average mean heart dose across the four cases presented is 0.68 Gy, 1.03 Gy, and 0.93 Gy, for the DPP, DPM and SPM cases respectively, while average contralateral breast/CW dose is 0.38 Gy, 0.58 Gy and 0.63 Gy. The V20 for the ipsilateral lung was, on average, reduced for the DPP expander at 14.3% across the 4 cases, while 18.8% and 17.0% for the DPM and SPM cases respectively.

CONCLUSION

The minimization of metal in the DPP expander allows for a significantly smaller beam avoidance region, in-turn enabling creation of a more homogeneous plan that can be potentially more robust while also reducing dose to surrounding OARs. Additional work is underway to expand the study to include a larger patient cohort, while also evaluating the impact of different robustness goals on OAR sparing.

The Anatomy and Evolution of a Quality Proton Therapy Plan Check Process: An Institutional Review

PURPOSE/OBJECTIVE(S)

There exists limited standard guidance from the physics community on the optimal way to ensure high-reliability proton therapy treatments. Following the guidance of AAPM MPPG 4a/4b and Task Groups 100/275, our institution has developed, implemented, and maintained strategies over the last five years to streamline workflows, enable more complex treatments, standardize procedures, and manage risk from slip or omission type errors in proton therapy. The evolution and success of that is observed from an institutional analysis of past operations data.

MATERIALS/METHODS

From the outset, standard care paths were used to sequence our workflows. These chains of tasks were well-defined and timestamped upon completion for every patient. Over five years, additional multiple interventions were implemented including clinical site standardization, checklists, root cause analysis, and failure mode identification. These were used to counter environmental variables with potential to disrupt timelines like growing patient volume, staffing demands, and new technique adoptions. These interventions went through extensive design, validation, and continuous improvement phases by the entire clinical team. Using ARIA v15.5, task timestamp data from the past 5 years were mined for physics proton-specific planning tasks and used to compute how early/late each task was in comparison to the expected due time. Average and standard deviation data were computed and used to create control charts. This coupled with the average time per year was cross-referenced with interventions to determine the impact on plan preparation timelines.

RESULTS

Over 5 years, the standard care path length reduced from 16 to 13 business days. The greatest impact on timelines was observed from standardizing clinical site guidelines. This minimized variability in task times and allowed the creation of checklists. By standardizing, we identified slack as well the need for new tasks. Overall, the average days late or early went from on-average late to on-average early (Table 1). The control charts indicate that many tasks stabilized in recent years. Furthermore, these interventions allowed us to safely increase the complexity of our proton plans using Monte Carlo dose calculations, 4D-optimization, robustness standards, repainting, and increased beams with no slippage in overall timelines.

CONCLUSION

Clinical site standardization, checklists, consistent process re-evaluation, and a team attitude of continuous improvement allowed us to improve control of our processes while simultaneously expanding treatment complexity and meeting the goals of our quality management plan.

BOLD fMRI response to direct stimulation (transcranial magnetic stimulation) of the motor cortex shows no decline with age

Previous studies using BOLD fMRI to examine age-related changes in cortical activation used tasks that relied on peripheral systems to activate the brain. They were unable to distinguish between alterations due to age-related changes in the periphery and actual changes in cortical physiology. Transcranial magnetic stimulation (TMS), which allows direct, noninvasive stimulation of cortical neurons, was interleaved with BOLD fMRI to study 6 young and 5 old subjects. Three different tasks were compared: direct stimulation by TMS, indirect active stimulation produced by a motor task, and indirect passive stimulation produced by hearing the TMS coil discharge. Direct neuronal stimulation by TMS produced similar fMRI signal increases in both groups, suggesting that cortical physiology itself may not necessarily decline with age.

The transcranial magnetic stimulation motor threshold depends on the distance from coil to underlying cortex: a replication in healthy adults comparing two methods of assessing the distance to cortex

Using transcranial magnetic stimulation (TMS), a handheld electrified copper coil against the scalp produces a powerful and rapidly oscillating magnetic field, which in turn induces electrical currents in the brain. The amount of electrical energy needed for TMS to induce motor movement (called the motor threshold [MT]), varies widely across individuals. The intensity of TMS is dosed relative to the MT. Kozel et al observed in a depressed cohort that MT increases as a function of distance from coil to cortex. This article examines this relationship in a healthy cohort and compares the two methods of assessing distance to cortex. Seventeen healthy adults had their TMS MT determined and marked with a fiducial. Magnetic resonance images showed the fiducials marking motor cortex, allowing researchers to measure distance from scalp to motor and prefontal cortex using two methods: 1) measuring a line from scalp to the nearest cortex and 2) sampling the distance from scalp to cortex of two 18-mm-square areas. Confirming Kozel's previous finding, we observe that motor threshold increases as distance to motor cortex increased for both methods of measuring distance and that no significant correlation exists between MT and prefontal cortex distance. Distance from TMS coil to motor cortex is an important determinant of MT in healthy and depressed adults. Distance to prefontal cortex is not correlated with MT, raising questions about the common practice of dosing prefontal stimulation using MT determined over motor cortex.

Motor cortex brain activity induced by 1-Hz transcranial magnetic stimulation is similar in location and level to that for volitional movement

RATIONALE AND OBJECTIVES

The relatively high temporal and spatial resolution of functional MR imaging was used to compare the blood oxygenation level dependent (BOLD) response associated with movement induced by transcranial magnetic stimulation (TMS) with that for a similar movement executed volitionally (VOL).

METHODS

Seven healthy adults were studied in a 1.5-T MR scanner. One hertz TMS at 110% of motor threshold was applied over the motor cortex for the thumb in 21-pulse trains in alternation with VOL every 63 seconds and interleaved with functional MR imaging.

RESULTS

BOLD increases in motor cortex associated with TMS and VOL movement were similar (2%-3%). Mean separation of their centers of activity was 3.7 + 1.9 mm (mean displacement: left/right = 0.3 +/- 4.1 mm; superior/inferior = 0.7 +/- 1.9 mm). There was no indication of supraphysiological brain activity.

CONCLUSIONS

Motor cortex BOLD response associated with thumb movement induced by 1-Hz TMS at 110% motor threshold is similar in both location and level to that caused by a similar movement executed volitionally.

A combined TMS/fMRI study of intensity-dependent TMS over motor cortex

BACKGROUND

Transcranial magnetic stimulation (TMS) allows noninvasive stimulation of neurons using time-varying magnetic fields. Researchers have begun combining TMS with functional imaging to simultaneously stimulate and image brain activity. Recently, the feasibility of interleaving TMS with functional magnetic resonance imaging (fMRI) was demonstrated. This study tests this new method to determine if TMS at different intensities shows different local and remote activation.

METHODS

Within a 1.5 Tesla (T) MRI scanner, seven adults were stimulated with a figure-eight TMS coil over the left motor cortex for thumb, while continuously acquiring blood oxygen level dependent (BOLD) echoplanar images. TMS was applied at 1 Hz in 18-second long trains delivered alternately at 110% and 80% of motor threshold separated by rest periods.

RESULTS

Though the TMS coil caused some artifacts and reduced the signal to noise ratio (SNR), higher intensity TMS caused greater activation than lower, both locally and remotely. The magnitude (approximately 3% increase) and temporal onset (2 to 5 sec) of TMS induced blood flow changes appear similar to those induced using other motor and cognitive tasks.

CONCLUSIONS

Though work remains in refining this potentially powerful method, combined TMS/fMRI is both technically feasible and produces measurable dose-dependent changes in brain activity.

Subcellular localization of the env-related glycoproteins in Friend erythroleukemia cells

A scheme was developed for the subcellular fractionation of murine erythroleukemia cells transformed by Friend leukemia virus. The subcellular localization of the env-related glycoproteins was determined by immune precipitation with antiserum against gp70, the envelope glycoprotein of the helper virus, followed by gel electrophoresis. In cells labeled for 2 h with [35S]methionine, the glycoprotein encoded by the defective spleen focus-forming virus, gp55SFFV, was found primarily in the nuclear fraction and in fractions containing dense cytoplasmic membranes such as endoplasmic reticulum. A similar distribution was noted for gp85env, the precursor to gp70. The concentration of viral glycoproteins in the nuclear fraction could not be accounted for by contamination with endoplasmic reticulum. In pulse-chase experiments, neither glycoprotein underwent major redistribution. However, labeled gp85env disappeared from intracellular membranes with a half-time of 30 min to 1 h, whereas labeled gp55SFFV was stable during a 2-h chase. In plasma membrane preparations with very low levels of contamination with endoplasmic reticulum, gp70 was the major viral env-related glycoprotein detected; a minor amount of gp55SFFV and no gp85env could be detected. The unexpected result of these experiments is the amount of viral glycoproteins found in the nuclear fraction. Presence of viral proteins in the nucleus could be relevant to the mechanism of viral leukemogenesis.