Mihaela Rosu

A fluence convolution method to account for respiratory motion in three-dimensional dose calculations of the liver: A Monte Carlo study

We describe the implementation of a fluence convolution method to account for the influence of superior-inferior (SI) respiratory induced motion on a Monte Carlo-based dose calculation of a tumor located in the liver. This method involves convolving the static fluence map with a function describing the SI motion of the liver-the motion function has been previously derived from measurements of diaphragm movement observed under fluoroscopy. Significant differences are noted between fluence-convolved and static dose distributions in an example clinical treatment plan; hot and cold spots (on the order of 25%) are observed in the fluence-convolved plan at the superior and inferior borders of the liver, respectively. This study illustrates that the fluence convolution method can be incorporated into Monte Carlo dose calculation algorithms to account for some of the effects of patient breathing during radiotherapy treatment planning, thus leading to more accurate dose calculations.

Accounting for center-of-mass target motion using convolution methods in Monte Carlo-based dose calculations of the lung.

We have applied convolution methods to account for some of the effects of respiratory induced motion in clinical treatment planning of the lung. The 3-D displacement of the GTV center-of-mass (COM) as determined from breath-hold exhale and inhale CT scans was used to approximate the breathing induced motion. The time-course of the GTV-COM was estimated using a probability distribution function (PDF) previously derived from diaphragmatic motion [Med. Phys. 26, 715-720 (1990)] but also used by others for treatment planning in the lung [Int. J. Radiat. Oncol., Biol., Phys. 53, 822-834 (2002); Med. Phys. 30, 1086-1095 (2003)]. We have implemented fluence and dose convolution methods within a Monte Carlo based dose calculation system with the intent of comparing these approaches for planning in the lung. All treatment plans in this study have been calculated with Monte Carlo using the breath-hold exhale CT data sets. An analysis of treatment plans for 3 patients showed substantial differences (hot and cold spots consistently greater than +/- 15%) between the motion convolved and static treatment plans. As fluence convolution accounts for the spatial variance of the dose distribution in the presence of tissue inhomogeneities, the doses were approximately 5% greater than those calculated with dose convolution in the vicinity of the lung. DVH differences between the static, fluence and dose convolved distributions for the CTV were relatively small, however, larger differences were observed for the PTV. An investigation of the effect of the breathing PDF asymmetry on the motion convolved dose distributions showed that reducing the asymmetry resulted in increased hot and cold spots in the motion convolved distributions relative to the static cases. In particular, changing from an asymmetric breathing function to one that is symmetric results in an increase in the hot/cold spots of +/- 15% relative to the static plan. This increase is not unexpected considering that the target spends relatively more time at inhale as the asymmetry decreases (note that the treatment plans were generated using the exhale CT scans).

The influence of beam model differences in the comparison of dose calculation algorithms for lung cancer treatment planning.

In this study, we show that beam model differences play an important role in the comparison of does calculated with various algorithms for lung cancer treatment planning. These differences may impact the accurate correlation of dose with clinical outcome. To accomplish this, we modified the beam model penumbral parameters in an equivalent path length (EPL) algorithm and subsequently compared the EPL doses with those generated with Monte Carlo (MC). A single AP beam was used for beam fitting. Two different beam models were generated for EPL calculations: (1) initial beam model (init_fit) and (2) optimized beam model (best_fit) , with parameters optimized to produce the best agreement with MC calculated profiles at several depths in a water phantom. For the 6 MV, AP beam, EPL(init_fit) calculations were on average within 2%/2 mm (1.4 mm max.) agreement with MC; the agreement for EPL(best_fit) was 2%/1.0 mm (1.3 mm max.) for EPL(best_fit). Treatment planning was performed using a realistic lung phantom using 6 and 15 MV photons. In all homogeneous phantom plans, EPL(best_fit) calculations were in better agreement with MC. In the heterogeneous 6 MV plan, differences between EPL(best_fit and init_fit) and MC were significant for the tumour. The EPL(init_fit), unlike the EPL(best_fit) calculation, showed large differences in the lung relative to MC. For the 15 MV heterogeneous plan, clinically important differences were found between EPL(best_fit or init_fit) and MC for tumour and lung, suggesting that the algorithmic difference in inhomogeneous cases, differences between EPL(best_fit) and MC for lung tissues were smaller compared to those between EPL(init_fit) and MC. Although the extent to which beam model differences impact the dose comparisons will be dependent upon beam parameters (orientation, field size and energy), and the size and location of the tumour, this study shows that failing to correctly account for beam model differences will lead to biased comparisons between dose algorithms. This may ultimately hinder our ability to accurately correlate dose with clinical outcome.

WE-C-ValA-02: The Impact of 4D Breathing Motion Effects Versus Tissue Heterogeneity in Lung Cancer Treatment Planning

Purpose: To investigate the relative magnitudes and clinical importances of the dosimetric effects related to 4D breathing motion and tissue heterogeneity for thoracic tumorstreatment planning.Methods: Scans were acquired at normal exhale/inhale breathing phases. The target was the union of the exhale and inhale GTVs, uniformly expanded by 5mm(ITV). Patients were planned with both AP/PA and 3‐D conformal plans using the exhale (“static”) dataset, assuming unit density, for 100±5% ITV dose coverage. Each of these plans was further used to calculate: (a) heterogeneous “static” dose; (b) homogeneous cumulative dose; (c) heterogeneous cumulative dose. The same number of MU were used for each of the calculations and was based on the homogeneous “static” plan. Cumulative dose distributions consisted of a time‐weighted sum of exhale and inhale doses. Doses were calculated using the DPM_MC code which includes secondary electrontransport for the heterogeneous computations. Results: Relative to unit‐density plans, tumor EUD, and lung NTCP increased in the heterogeneity corrected plans; primarily due to the reduced beam attenuation through lungs and the larger than coin‐size tumors investigated. In comparing 4D cumulative dose plans with static plans, clinical EUD and NTCP estimates were relatively unchanged. The insignificant tumor EUD change was a consequence of good target design, while the small lung NTCP change was due to its large volume effect. Accounting for tissue heterogeneity resulted in average changes of 10% in MLD. Accounting for 4D breathing motion effects resulted in <1% changes in MLD from the static value. The magnitude of these effects was not correlated with the dose distribution conformality. Conclusions: In this study we found that tissue heterogeneity effects are likely to have a larger clinical significance on tumor (if ITV is properly designed) and normal lung clinical treatment evaluation metrics than occurs with 4‐D respiratory‐induced changes. Supported by P01‐CA59827, R01‐CA106770.

SU-FF-T-76: Implementation and Initial Testing of a Monte Carlo Based Algorithm for IMRT Inverse Treatment Planning

Purpose: To report on the implementation of a Monte Carlo(MC) based algorithm and to compare this system with a convolution/superposition‐based algorithm (CS) for IMRT inverse planning. Method and Materials: The DPM MC code was modified using a fluence matrix approach to perform beamlet calculations for IMRT planning. The code was integrated within our in‐house inverse treatment planning system and compared with the TPS (CS) algorithm. Initial testing involved the computation of 6 MV beamlet depth doses for 1×1, 2×2 and 10×10 (100, 1 cm beamlets) in a water phantom. MC and CS calculations were then performed for an example lung treatment plan to examine dosimetric differences between these algorithms. MC statistical uncertainties were on average less than 2% (in the depth doses) for all beamlet calculations. Optimization of beamlet doses is carried out using simulated annealing with quadratic cost functions derived from our clinical protocols. Results: Beamlet depth doses calculated with MC and CS are in good absolute agreement for field sizes larger than 2×2 cm2. Significant differences exist for 1×1 beamlets because CS is unable to accurately model lateral electron transport. For the example lung plan, much smaller differences were found. This is likely due to the fact that with larger field sizes (∼10×10 cm in the example), effects of lateral electron scattering are much less pronounced. Conclusion: We have implemented a fluence matrix method to perform MC‐based beamlet calculations for IMRT planning. Initial testing for an example lung plan and field sizes larger than 2×2, revealed good agreement between MC and CS. However, larger differences were found for 1×1 beamlets due to lateral electron transport issues. Testing is currently being performed for a variety of treatment plans, spanning a range of field sizes to thoroughly investigate dosimetric differences between MC and CS in IMRT planning. Supported by NIH P01‐CA59827

SU‐FF‐J‐19: How Much 4D Data Is Needed for Estimation of Dose Distribution Evaluation Metrics at the Time of Planning?

Purpose: To investigate the number of intermediate states required to adequately approximate the cumulative dose to deforming/moving thoracic anatomy. Method and Materials: CT scans at exhale were registered to images and/or simulated data at inhale and 4 or more intermediate breathing states for several lung cancer patients using B-spline transformations. Doses to each state were computed using the DPM Monte-Carlo code and dose was accumulated for scoring on the exhale anatomy via the transformation matrices for each state and time weighting factors. Cumulative doses were estimated using increasing numbers of intermediate states and compared to simpler scenarios such as a “2-state” model which used only the exhale and inhale datasets (as these have the highest time weighting coefficients). Dose distributions for each modeled state as well as the cumulative doses were assessed using DVHs and several treatment evaluation metrics such as mean lung dose, NTCP and gEUD). Results: Although significant “point dose” differences can exist between each breathing state, the differences decrease when cumulative doses are considered, and can become less significant yet in terms of evaluation metrics depending upon clinical endpoint. For example, differences between a “2-state” and a “6-state” cumulative dose distribution are often within a few percent of each other and can have no significant clinical impact on treatment metrics for the lung itself as it is a large volume effect organ. However, the use of more intermediate states is sometimes required to properly estimate doses to other adjacent organs at risk such as the esophagus. Conclusion: This ongoing study suggests that for certain “clinical” endpoints it may only be necessary to properly give weight and accumulate the doses from a few well separated states (having the highest probability of occurrence) to achieve satisfactory predictions of the results of accumulating dose to the distorting anatomy. Supported by NIH P01CA59827.