Histogram reduction method for calculating complication probabilities for three-dimensional treatment planning evaluations

Proton therapy for pediatric cranial tumors: preliminary report on treatment and disease-related morbidities

PURPOSE

Accelerated protons were used in an attempt to limit treatment-related morbidity in children with tumors in or near the developing brain, by reducing the integral dose to adjacent normal tissues.

METHODS AND MATERIALS

Children treated with protons at Loma Linda University Medical Center between August 1991 and December 1994 were analyzed retrospectively. Twenty-eight children, aged 1 to 18 years, were identified as at risk for brain injury from treatment. Medical records, physical examinations, and correspondence with patients, their parents, and referring physicians were analyzed. The investigators tabulated post-treatment changes in pre-treatment signs and symptoms and made judgments as to whether improvement, no change, or worsening related to disease or treatment had supervened. Magnetic resonance images were correlated with clinical findings and radiographic impressions were tabulated.

RESULTS

Follow-up ranged from 7 to 49 months (median 25 months). Four instances of treatment-related morbidity were identified. Forty-one instances of site-specific, disease-related morbidity were identified: 15 improved or resolved and 26 remained unchanged after treatment. Four patients had radiographic evidence of local failure. Three of these patients, including two with high-grade glioma, have died.

CONCLUSION

Early treatment-related morbidity associated with proton therapy is low. Tumor progression remains a problem when treating certain histologies such as high-grade glioma. Escalating the dose delivered to target volumes may benefit children with tumors associated with poor rates of local control. Long-term follow-up, including neurocognitive testing, is in progress to assess integral-dose effects on cognitive, behavioral and developmental outcomes in children with cranial tumors.

Dose-volume histograms for bladder and rectum

PURPOSE

A careful examination of the foundation upon which the concept of the Dose-Volume Histogram (DVH) is built, and the implications of this set of parameters on the clinical application and interpretation of the DVH concept has not been conducted since the introduction of DVHs as a tool for the quantitative evaluation of treatment plans. The purpose of the work presented herein is to illustrate problems with current methods of implementing and interpreting DVHs when applied to hollow anatomic structures such as the bladder and rectum.

METHODS AND MATERIALS

A typical treatment plan for external beam irradiation of a patient with prostate cancer was chosen to provide a data set from which DVH curves for both the bladder and rectum were calculated. The two organs share the property of being shells with contents that are of no clinical importance. DVHs for both organs were computed using a solid model and using a shell model. Typical treatment plans for prostate cancer were used to generate DVH curves for both models. The Normal Tissue Complication Probability (NTCP) for these organs is discussed in this context.

RESULTS

For an eight-field conformal treatment plan of the prostate, a bladder DVH curve generated using the shell model is higher than the corresponding curve generated using the solid model. The shell model also has a higher NTCP. A six-field conformal treatment plan also results in a higher DVH curve for the shell model. A treatment plan consisting of bilateral 120-degree arcs, results in a higher DVH curve for the shell model, as well as a higher NTCP.

CONCLUSION

The DVH concept currently used in evaluation of treatment plans is problematic because current practices of defining exactly what constitutes "bladder" and "rectum." Commonly used methods of tracing the bladder and rectum imply use of a solid structure model for DVHs. In reality, these organs are shells and the critical structure associated with NTCP is obviously and indisputably the shell, as opposed to its contents. Treatment planning algorithms for DVH computation should thus be modified to utilize the shell model for these organs.

Estimation of the spatial distribution of target cells for radiation pneumonitis in mouse lung

PURPOSE

Recent studies of Liao et al. and Travis et al. have demonstrated that irradiation of cross-sectional partial volumes contiguous to the base of mouse lung produces a higher incidence of pneumonitis than irradiation of equally sized subvolumes contiguous to the apex. One interpretation of this finding is that the critical target cells for pneumonitis are not distributed uniformly throughout the lung. The purpose of the present study was to test this interpretation by obtaining an estimate of the spatial distribution of the critical target cells for pneumonitis in mouse lung, based on the partial-volume data.

METHODS AND MATERIALS

A mathematical model was derived describing the probability of radiation pneumonitis as a function of dose, volume of lung irradiated, and location in the lung of the irradiated subvolume. The model includes a nonparametric description of the spatial target-cell distribution in lung, to be estimated from data. The model was fitted to the lethality data of Liao et al. and Travis et al. to obtain estimates of the proportion of target cells contained in each of various subvolumes of the lung.

RESULTS

The results indicate that the critical target cells in mouse lung are concentrated in the base and, to a somewhat lesser extent, in the apex of the lung, with fewer cells in the middle region. The estimated spatial distribution of target cells in mouse lung agrees well with the distribution of alveoli, whose concentration in the midlung is limited by the presence of the major conducting airways.

CONCLUSION

Heterogeneity in the spatial distribution of critical target cells in normal tissue implies that the complication probability (NTCP) depends on the location in the organ of the irradiated subvolume, as well as on radiation dose and subvolume size. Calculations using an NTCP model for mouse lung indicate that irradiation of equal subvolumes of lung with the same dose can lead to complication probabilities covering the full range from 0 to 100%, depending only on the location in the lung of the irradiated subvolume.

Spatial heterogeneity of the volume effect for radiation pneumonitis in mouse lung

PURPOSE

In a previous study to determine the effect of partial volume irradiation on damage and morbidity from pneumonitis in mouse lung, a critical determinant of the volume effect was the spatial location of the irradiated subvolume within the lung. The goals of the present study were to (a) define the dose-volume effect curves for radiation pneumonitis in mouse lung, (b) define the threshold volume, and (c) further investigate the spatial heterogeneity of the radiosensitivity of mouse lung.

METHODS AND MATERIALS

Eight fractional volumes ranging from 94% to 17% of the lungs of C3Hf/Kam mice were irradiated with single doses ranging from 12 to 22 Gy, depending on the volume irradiated. The fractional volumes irradiated were determined from computed tomographic scans of mouse lung. To determine the effect of location of irradiated subvolume, equivalent volumes in the base and the apex were irradiated by shielding the prescribed adjacent volume in the apex or base respectively. Dose-response curves of breathing rate at 22 weeks and lethality at 28 weeks were constructed for each subvolume irradiated in the apex or base and fitted by logit analysis, and ED50s and LD50s with 95% confidence limits obtained, respectively. Lungs from dead mice or mice sacrificed when moribund were examined for histologic signs of pneumonitis.

RESULTS

Irradiation of any of the eight subvolumes in the base yielded a consistently lower isoeffect dose for both assays of radiation pneumonitis than if the same irradiated subvolume was located in the apex. Plots of isoeffect dose for breathing rate as a function of subvolume irradiated in the base or apex showed that these curves were not linear but exhibited a plateau between irradiated volumes of 70% and 80% in both the apex and base. A similar curve was obtained for lethality and volume irradiated in the base. A threshold volume, i.e., irradiation of that volume that should produce no changes in breathing rate or mortality, was dependent on the location of the irradiated subvolume.

CONCLUSION

The response of mouse lung to partial volume irradiation is heterogeneous and is critically dependent on the specific location of the irradiated subvolume in the lung, i.e., a given subvolume in the base is consistently more sensitive than the same subvolume in the apex using either breathing rate or lethality as assays of radiation pneumonitis. We suggest that this heterogeneity is due to the anatomy of the tracheobronchial tree, i.e., to the distribution of non-gas exchange-conducting airways in the irradiated volume. These data have implications for the modeling of dose-volume effects in the lung and the prediction of normal tissue complication probabilities for radiation pneumonitis in humans.

Dose-volume complication analysis for visual pathway structures of patients with advanced paranasal sinus tumors

PURPOSE

The purpose of the present work was to relate dose and volume information to complication data for visual pathway structures in patients with advanced paranasal sinus tumors.

METHODS AND MATERIALS

Three-dimensional (3D) dose distributions for chiasm, optic nerve, and retina were calculated and analyzed for 20 patients with advanced paranasal sinus malignant tumors. 3D treatment planning with beam's eye view capability was used to design beam and block arrangements, striving to spare the contralateral orbit (to lessen the chance of unilateral blindness) and frequently the ipsilateral orbit (to help prevent bilateral blindness). Point doses, dose-volume histogram analysis, and normal tissue complication probability (NTCP) calculations were performed. Published tolerance doses that indicate significant risk of complications were used as guidelines for analysis of the 3D dose distributions.

RESULTS

Point doses, percent volume exceeding a specified published tolerance dose, and NTCP calculations are given in detail for patients with complications versus patients without complications. Two optic nerves receiving maximum doses below the published tolerance dose sustained damage (mild vision loss). Three patients (of 13) without optic nerve sparing and/or chiasm sparing had moderate or severe vision loss. Complication data, including individual patient analysis to estimate overall risk for loss of vision, are given.

CONCLUSION

3D treatment planning techniques were used successfully to provide bilateral sparing of the globe for most patients. It was more difficult to spare the optic nerves, especially on the ipsilateral side, when prescription dose exceeded the normal tissue tolerance doses. NTCP calculations may be useful in assessing complication risk better than point dose tolerance criteria for the chiasm, optic nerve, and retina. It is important to assess the overall risk of blindness for the patient in addition to the risk for individual visual pathway structures.

Predicting late rectal complications following prostate conformal radiotherapy using biologically effective doses and normalized dose-surface histograms

A model to predict the late normal tissue complication probability (NTCP) of the rectum following conformal therapy is described. The model evaluates the biological consequence of inhomogeneities in the physical dose by computing dose histograms of the biologically effective dose to the surface of the rectum for a given fractionation scheme. A method of normalizing the surface area of the rectum is employed so that the predicted NTCP is independent of the differing cross-sectional size of sections of the rectum, ensuring the NTCP is dependent only on the dose delivered to sensitive rectal tissues. The model has been used to assess severe late rectal complications and the milder RTOG grades 2 and 3 reactions. This model was found to predict severe toxicity levels of 1.7 +/- 0.6% for an accelerated treatment of 50 Gy in 16 fractions commonly employed at this centre. This result lies between the severe toxicities predicted for 60 and 62 Gy delivered in 2 Gy fractions. The model predicts that the average NTCP for severe late effects for nine prostate patients becomes greater than 5% with a fractionation scheme of 70 Gy in 35 fractions, for the four fields treatment. The effects of not treating all fields at each therapy session on rectal toxicity were also investigated. Biologically effective dose-surface histograms show that the dose to the lower surface of the rectum is increased by not treating all fields at each therapy session, but the predicted differences in rectal NTCP are negligible.

Treatment portals for elective radiotherapy of the neck: an inventory in The Netherlands

PURPOSE

To assess the variation in and the three-dimensional dosimetric consequences of treatment portals for elective neck irradiation.

MATERIALS AND METHODS

Radiation oncologists (n = 16) from all major Head and Neck Co-operative Groups in The Netherlands (n = 11) were asked to delineate treatment portals on a lateral and an anterior simulation film in case of elective neck irradiation for a T3N0 tumour of the supraglottic larynx and a T2N0 tumour of the mobile tongue. In addition, they had to define their target, i.e. which parts of the neck nodal regions they would choose to irradiate electively for these particular tumour sites. Subsequently, treatment portals were compared and evaluated using CT-data and a 3-dimensional (3D) treatment planning system.

RESULTS

Significant variations were found in the shapes and sizes of the applied treatment techniques and portals. Also, among radiation oncologists who elected to irradiate the same lymph node regions, a significant variation in the delineated treatment portals was observed. As a consequence, substantial variations in treated volumes and in calculated normal tissue complication probabilities (NTCPs) for the parotid- and submandibular glands were observed.

CONCLUSION

For the tumour sites studied there appears to be a lack of standardisation in the areas of the neck to be irradiated electively. The observed differences may have consequences for the ultimate failure rate and particularly with regard to the side effects, e.g. the degree of xerostomia. It is argued that in the near future a more precise three-dimensional definition on CT of the lymph node regions in the neck might allow for a better standardisation of the treatment portals and, in addition, for the development and application of conformal radiotherapy techniques for optimal sparing of the critical normal tissues (e.g. parotid- and submandibular glands) with maximum tumour control probability.

Dose escalation for non-small cell lung cancer using conformal radiation therapy

PURPOSE

Improved local control of non-small cell lung cancer (NSCLC) may be possible with an increased dose of radiation. Three-dimensional radiation treatment planning (3D RTP) was used to design a radiation therapy (RT) dose escalation trial, where the dose was determined by (a) the effective volume of normal lung irradiated, and (b) the estimated risk of a complication. Preliminary results of this trial were reviewed.

METHODS AND MATERIALS

A graph of the iso-normal tissue complication probability (NTCP) levels associated with a dose and effective volume (V(eff)) was derived, using normal tissue parameters derived from the literature. This led to a dose escalation schema, where patients were sorted into 1 of 5 treatment bins, determined by the V(eff) of the best possible treatment plan. The starting doses ranged from 63 to 84 Gy. Each treatment bin was then escalated separately, as in Phase I dose escalation fashion, with Grade > or = 3 radiation pneumonitis defined as dose limiting. To allow for dose escalation, we required patient follow-up to be > or = 6 months for at least three patients. 3D treatment planning was used to irradiate only the radiographically abnormal areas, with 2.1 Gy (corrected for lung inhomogeneity)/day. Clinically uninvolved lymph nodes were not treated prophylactically.

RESULTS

A total of 48 NSCLC patients have been treated (Stage I/II: 18 patients; Stage III: 28 patients; mediastinal recurrence postsurgery: 2 patients). No radiation pneumonitis has been observed in the 30 patients currently evaluable beyond the 6-month time point. All treatment bins have been escalated at least once. Current doses in the five treatment bins are 69.3, 69.3, 75.6, 84, and 92.4 Gy. None of the 15 evaluable patients in any bin with > or = 30% NTCP experienced clinical radiation pneumonitis, implying that the actual risk is < 20% (beta error rate 5%). Despite the observation of the clinically negative lymph nodes at high risk, there has been no failure in the untreated mediastinum as the sole site of first failure. Three of 10 patients receiving > or = 84 Gy have had biopsy proven residual or locally recurrent disease.

CONCLUSION

Successful dose escalation in a volume-dependent organ can be performed using this technique. By incorporating the effective volume of irradiated tissue, some patients have been treated to a total dose of radiation over 50% higher than traditional doses. The literature-derived parameters appear to overestimate pneumonitis risk with higher volumes. There has been no obvious negative effect due to exclusion of elective lymph node radiation. When completed, this trial will have determined the maximum tolerable dose of RT as a single agent for NSCLC and the appropriate dose for Phase II investigation.

Tumor control probability of nasopharyngeal carcinoma: a comparison of different mathematical models

PURPOSE

Radiation dose and tumor volume are factors known to affect the local control of a given type of tumor. Local tumor control is a major factor to consider when a treatment plan is evaluated. This article reports the correlation between tumor control probability, dose, and volume in a retrospective study of 142 patients with nasopharyngeal carcinoma (NPC).

METHODS AND MATERIALS

The tumor volume was outlined and calculated from a computed tomographic scan. Patients were categorized according to tumor volume and radiation dose received in treatment. Local control rate was calculated for each category by the Kaplan-Meier method. Mathematical models were fitted to correlate the local control rate, dose, and volume. Both empirical and mechanistic approaches were attempted; the former included logistic models with two and three parameters, and the latter, the formulation from Brenner and Bentzen with a radiobiological basis.

RESULTS

Brenner's model estimated alpha at 0.041 Gy(-1) with 95% confidence limits (-0.032, 0.113) Gy(-1). The volume dependent constant h was estimated at 0.160 cm(-3) with 95% confidence limits (-0.729, 1.048) cm(-3). The Pearson correlation coefficient was 0.64. The magnitude and sign of the fitted parameters were reasonable and consistent with reported clinical experience. The other models were fitted with slightly better goodness of fit (r = 0.65 - 0.68), but with less interpretable parameters.

CONCLUSION

Brenner's model is considered appropriate for a description of the dose and volume effect on the local control of the NPC. It could be used in combination with normal tissue complication probability for treatment plan evaluation to optimize treatment results.

Number and orientations of beams in intensity-modulated radiation treatments

The fundamental question of how many equispaced coplanar intensity-modulated photon beams are required to obtain an optimum treatment plan is investigated in a dose escalation study for a typical prostate tumor. Furthermore, optimization of beam orientations to improve dose distributions is explored. A dose-based objective function and a fast gradient technique are employed for optimizing the intensity profiles (inverse planning). An exhaustive search and fast simulated annealing techniques (FSA) are used to optimize beam orientations. However, to keep computation times reasonable, the intensity profiles for each beam arrangement are still optimized using inverse planning. A pencil beam convolution algorithm is employed for dose calculation. All calculations are performed in three-dimensional (3D) geometry for 15 MV photons. DVHs, dose displays, TCP, NTCP, and biological score functions are used for evaluation of treatment plans. It is shown that for the prostate case presented here, the minimum required number of equiangular beams depends on the prescription dose level and ranges from three beams for 70 Gy plans to seven to nine beams for 81 Gy plans. For the highest dose level (81 Gy), beam orientations are optimized and compared to equiangular spaced arrangements. It is shown that (1) optimizing beam orientations is most valuable for a small numbers of beams (< or = 5) and the gain diminishes rapidly for higher numbers of beams; (2) if sensitive structures (for example rectum) are partially enclosed by the target volume, beams coming from their direction tend to be preferable, since they allow greater control over dose distributions; (3) while FSA and an exhaustive search lead to the same results, computation times using FSA are reduced by two orders of magnitude to clinically acceptable values. Moreover, characteristics of and demands on biology-based and dose-based objective functions for optimization of intensity-modulated treatments are discussed.

Functional dose-volume histograms for functionally heterogeneous normal organs

Functional dose-volume histograms are proposed as an extension of the conventional dose-volume histograms, for quantitative assessment of three-dimensional radiation dose coverage of functionally heterogeneous normal organs. Examples are given to illustrate possible applications of this approach to the treatment of a brain tumour or a lung tumour, in which cases the distribution of the normal organ function can be obtained from functional dose-volume modalities. It is shown that a significant difference exists between the functional dose-volume histograms and the conventional dose-volume histograms when the normal organ function is non-uniformly distributed within the organ. Utilization of functional dose-volume histograms as the input for the calculation of normal tissue complication probabilities is discussed for different normal tissue structures.

Conformal irradiation of the prostate: estimating long-term rectal bleeding risk using dose-volume histograms

PURPOSE

Dose-volume histograms (DVHs) may be very useful tools for estimating probability of normal tissue complications (NTCP), but there is not yet an agreed upon method for their analysis. This study introduces a statistical method of aggregating and analyzing primary data from DVHs and associated outcomes. It explores the dose-volume relationship for NTCP of the rectum, using long-term data on rectal wall bleeding following prostatic irradiation.

METHODS AND MATERIALS

Previously published data were reviewed and updated on 41 patients with Stages T3 and T4 prostatic carcinoma treated with photons followed by perineal proton boost, including dose-volume histograms (DVHs) of each patient's anterior rectal wall and data on the occurrence of postirradiation rectal bleeding (minimum FU > 4 years). Logistic regression was used to test whether some individual combination of dose and volume irradiated might best separate the DVHs into categories of high or low risk for rectal bleeding. Further analysis explored whether a group of such dose-volume combinations might be superior in predicting complication risk. These results were compared with results of the "critical volume model," a mathematical model based on assumptions of underlying radiobiological interactions.

RESULTS

Ten of the 128 tested dose-volume combinations proved to be "statistically significant combinations" (SSCs) distinguishing between bleeders (14 out of 41) and nonbleeders (27 out of 41), ranging contiguously between 60 CGE (Cobalt Gray Equivalent) to 70% of the anterior rectal wall and 75 CGE to 30%. Calculated odds ratios for each SSC were not significantly different across the individual SSCs; however, analysis combining SSCs allowed segregation of DVHs into three risk groups: low, moderate, and high. Estimates of probabilities of normal tissue complications (NTCPs) based on these risk groups correlated strongly with observed data (p = 0.003) and with biomathematical model-generated NTCPs.

CONCLUSIONS

There is a dose-volume relationship for rectal mucosal bleeding in the region between 60 and 75 CGE; therefore, efforts to spare rectal wall volume using improved treatment planning and delivery techniques are important. Stratifying dose-volume histograms (DVHs) into risk groups, as done in this study, represents a useful means of analyzing empirical data as a function of hetereogeneous dose distributions. Modeling efforts may extend these results to more heterogeneous treatment techniques. Such analysis of DVH data may allow practicing clinicians to better assess the risk of various treatments, fields, or doses, when caring for an individual patient.

The "critical volume tolerance method" for estimating the limits of dose escalation during three-dimensional conformal radiotherapy for prostate cancer

PURPOSE

To describe the "Critical Volume Tolerance" (CVT) method for defining normal tissue tolerance during 3D-based dose escalation studies for prostate cancer.

METHODS AND MATERIALS

The CVT method predicts the tolerance to radiation for "in series"-type functional units based on the assumption that tolerance depends on a critical threshold "low-volume high-dose region." The data used for describing this model were generated from 3D analysis of randomly selected patients with prostate cancer. Commonly used coplanar four-and six-field conformal (SFC) techniques were chosen as the comparison techniques. For purposes of comparison, rectal tolerance was assumed to be reached following whole pelvic irradiation using a four-field box technique to 50 Gy, followed by a conedown boost to 70 Gy using bilateral 9 x 9 cm 120 degree arcs as popularized by investigators from Stanford University (SUH).

RESULTS

Based on the average dose volume histograms for the patients studied, the maximum safe increase in dose for the SFC technique compared to the SUH technique, would be 10% if 30% of the rectal volume was the critical dose limiting volume (CVT = 30%), 5% if the CVT = 10%, or greater than 20% if the CVT = 40%. Commonly used four-field conformal techniques would not be expected to allow significant escalation of the dose without increasing the risk of complications.

CONCLUSIONS

The CVT method is relatively simple, and data generated based on it can be used to support normal tissue complication probability equations. The CVT method can be verified or modified as partial tolerance data become available. Based on the CVT model, sophisticated treatment techniques should allow a modest increase in the total dose of radiation delivered to the prostate without an increase in late complications.

Evaluation and scoring of radiotherapy treatment plans using an artificial neural network

PURPOSE

The objective of this work was to demonstrate the feasibility of using an artificial neural network to predict the clinical evaluation of radiotherapy treatment plans.

METHODS AND MATERIALS

Approximately 150 treatment plans were developed for 16 patients who received external-beam radiotherapy for soft-tissue sarcomas of the lower extremity. Plans were assigned a figure of merit by a radiation oncologist using a five-point rating scale. Plan scoring was performed by a single physician to ensure consistency in rating. Dose-volume information extracted from a training set of 511 treatment plans on 14 patients was correlated to the physician-generated figure of merit using an artificial neural network. The neural network was tested with a test set of 19 treatment plans on two patients whose plans were not used in the training of the neural net.

RESULTS

Physician scoring of treatment plans was consistent to within one point on the rating scale 88% of the time. The neural net reproduced the physician scores in the training set to within one point approximately 90% of the time. It reproduced the physician scores in the test set to within one point approximately 83% of the time.

CONCLUSIONS

An artificial neural network can be trained to generate a score for a treatment plan that can be correlated to a clinically-based figure of merit. The accuracy of the neural net in scoring plans compares well with the reproducibility of the clinical scoring. The system of radiotherapy treatment plan evaluation using an artificial neural network demonstrates promise as a method for generating a clinically relevant figure of merit.

Head and neck cancer

This synthesis of the literature on radiotherapy for head and neck cancer is based on 424 scientific articles, including 3 meta-analyses, 38 randomized studies, 45 prospective studies, and 246 retrospective studies. These studies involve 79174 patients. The literature review shows that radiotherapy, either alone or in combination with surgery, plays an essential role in treating head and neck cancers. When tumors are localized, many tumor patients can be cured by radiotherapy alone and thereby maintain full organ function (1, 2). Current technical advancements in radiotherapy offer the potential for better local tumor control with lower morbidity (3). This, however, will require more sophisticated dose planning resources. To further improve treatment results for advanced tumors, other fractionation schedules, mainly hyperfractionation, should be introduced (5). This mainly increases the demands on staff resources for radiotherapy. The combination of radiotherapy and chemotherapy should be subjected to further controlled studies involving a sufficiently large number of patients (4, 5). Interstitial treatment (in the hands of experienced radiotherapists) yields good results for selected cancers. The method should be more generally accessible in Sweden. Intraoperative radiotherapy should be targeted for further study and development.

Advanced interactive planning techniques for conformal therapy: high level beam descriptions and volumetric mapping techniques

PURPOSE

To aid in design of conformal radiation therapy treatment plans involving many conformally shaped fields, this work investigates the use of two methodologies to enhance the ease of interactive treatment planning: high-level beam constructs and beam's-eye view volumetric mapping.

METHODS AND MATERIALS

High-performance computer graphics running on various workstations using a graphical visualization system (AVS) have been used in this work. Software specific to this application has been written in standard FORTRAN and C languages. A new methodology is introduced by defining radiation therapy "fields" to be composed of multiple beam "segments." Fields can then be defined as higher-level entities such as arcs, cones, and other shapes. A "segmental cone" field, for example, is defined by a symmetry axis and a cone angle, and can be used to rapidly place a series of beam segments that converge at the target volume, while reducing the degree of overlap elsewhere. A new beam's-eye view (BEV) volumetric mapping technique is presented to aid in selecting the placement of conformal radiation fields. With this technique, the relative average dose within an organ of interest is calculated for a sampling of isocentric, conformally shaped beams and displayed either as a "globe," which can be combined with the display of anatomical surfaces, or as a two-dimensionally mapped projection. The dose maps from multiple organs can be generated, stacked, or composited with relative weightings to aid in the placement of fields that minimize overlap with critical structures.

RESULTS

The use of these new methodologies is demonstrated for prostate and lung treatment sites and compared to conventional planning techniques.

DISCUSSION

The use of many beams for conformal treatment delivery is difficult with current interactive planning. The use of high-level beam constructs provides a means to quickly specify, place, and configure multiple beam arrangements. The BEV volumetrics aids in the placing of fields, which minimize involvement with critical normal tissues.

CONCLUSIONS

Early experience with the new methodologies suggest that the new methods help to enhance (or at least speed up) the ability of a treatment planner to create optimal radiation treatment field arrangements.

Variation in volumes, dose-volume histograms, and estimated normal tissue complication probabilities of rectum and bladder during conformal radiotherapy of T3 prostate cancer

PURPOSE

To determine the pattern of changes of rectum and bladder structures during conformal therapy of T3 prostate cancer and the impact of these changes on the accuracy of the dose-volume histograms (DVHs) and normal tissue complication probabilities (NTCPs) of these organs, based on the planning computed tomography (CT) scan only.

METHODS AND MATERIALS

For 11 T3 prostate cancer patients treated with conformal therapy, three repeat CT scans were made in Weeks 2, 4, and 6 of the treatment. The bony anatomy was aligned with the planning CT scan, using three dimensional (3D) chamfer matching. The internal and external surfaces of rectum and bladder were contoured in each scan. Three volumes were calculated for each organ: solid organ (including filling), filling, and wall volume. DVHs and NTCPs were calculated for all structures.

RESULTS

The solid organ and filling volumes varied considerably between patients and within a patient and they decreased with increasing treatment time. The largest patient variation was seen for patients with large initial filling volumes. The variations of rectum and bladder wall volumes during treatment were 9 and 17% (1 standard deviation (SD)), respectively, with no time trend. The changes of the high dose (> 80 and 90% of the prescribed dose) volumes of the rectum in response to rectum filling differences were proportional to the whole rectum volume changes. The variation of the high-dose rectum wall volume was relatively small (14%, 1 SD). As a result, the NTCPs of rectum and rectum wall were the same overall and the variation of the NTCPs during treatment was about 14% (1 SD) and not correlated with rectum filling. The variation of the high-dose bladder volumes (about 14%, 1 SD) was smaller than the variation of the whole bladder volumes (30%, 1 SD). The high-dose bladder wall volume decreased significantly due to wall distention as the bladder filling increased. As a result of this complex pattern, the variation of NTCPs of bladder (85%, 1 SD) and bladder wall (88%, 1 SD) during treatment was large and significantly correlated with bladder filling.

CONCLUSIONS

The planning CT scan overestimates rectum and bladder filling during treatment. Furthermore, the variation of filling is so large that only the wall structures have relatively constant volumes during treatment. For the rectum wall, the DVHs and NTCPs, as estimated from the initial scan, are representative for the whole treatment, because no correlation was seen between these parameters and organ filling. For the bladder wall, however, such a correlation was present and consequently, the initial bladder wall DVHs and NTCPs can only be representative for the whole treatment, if the bladder filling can be kept reasonably constant during treatment.

Estimation of pneumonitis risk in three-dimensional treatment planning using dose-volume histogram analysis

PURPOSE

Investigations to study correlations between the estimations of biophysical models in three dimensional (3D) treatment planning and clinical observations are scarce. The development of clinically symptomatic pneumonitis in the radiotherapy of thoracic malignomas was chosen to test the predictive power of Lyman's normal tissue complication probability (NTCP) model for the assessment of side effects for nonuniform irradiation.

METHODS AND MATERIALS

In a retrospective analysis individual computed-tomography-based 3D dose distributions of a random sample of 46/20 patients with lung/esophageal cancer were reconstructed. All patients received tumor doses between 50 and 60 Gy in a conventional treatment schedule. Biological isoeffective dose-volume histograms (DVHs) were used for the calculation of complication probabilities after applying Lyman's and Kutcher's DVH-reduction algorithm. Lung dose statistics were performed for single lung (involved ipsilateral and contralateral) and for the lung as a paired organ.

RESULTS

In the lung cancer group, about 20% of the patients (9 out of 46) developed pneumonitis 3-12 (median 7.5) weeks after completion of radiotherapy. For the majority of these lung cancer patients, the involved ipsilateral lung received a much higher dose than the contralateral lung, and the pneumonitis patients had on average a higher lung exposure with a doubling of the predicted complication risk (38% vs. 20%). The lower lung exposure for the esophagus patients resulted in a mean lung dose of 13.2 Gy (lung cancer: 20.5 Gy) averaged over all patients in correlation with an almost zero complication risk and only one observed case of pneumonitis (1 out of 20). To compare the pneumonitis risk estimations with observed complication rates, the patients were ranked into bins of mean ipsilateral lung dose. Particularly, in the bins with the highest patient numbers, a good correlation was achieved. Agreement was not reached for the lung functioning as a paired organ.

CONCLUSIONS

Realistic assessments for the prediction of radiation-induced pneumonitis seem to be possible. In this respect, the implementation of DVH-analysis in 3D planning could be a helpful tool for the evaluation of treatment plans.

Damage and morbidity from pneumonitis after irradiation of partial volumes of mouse lung

PURPOSE

The aims of this study were to: (a) define the relationship of dose and volume irradiated to damage and morbidity in mouse lung, (b) determine the threshold volume for morbidity after partial lung irradiation; and (c) determine whether the response to radiation of mouse lung is independent of the region irradiated.

METHODS AND MATERIALS

C3Hf/Kam female mice were used in this study. The fractional volume of the lung to be irradiated was determined by two methods, weights and computed tomography (CT) scanning. Two experiments were performed to define the volume effect and to determine whether the response of the mouse lung to radiation was homogeneous. In the first experiment, single doses of x-rays ranging from 12 to 20 Gy were given to partial volumes of 84%, 70%, and 40% including the base, 50%, 33%, and 17% including the apex, to 43% in the middle, and to the sum of 57% as 17% in the apex and 40% in the base. In the second experiment, the same volumes of 50% and 70-75% in the apex and base of the lung were irradiated with single doses ranging from 12-19.25 Gy. Morbidity from radiation pneumonitis was quantitated by two end points, breathing rate and lethality between 12 and 32 weeks after irradiation. Damage was assessed by histopathological evidence of pneumonitis.

RESULTS

Clear well-defined dose-response curves were obtained for both breathing rate and lethality after all volumes irradiated. There was a clear volume-dependent shift of the dose-response curves for breathing rate and lethality at 28 weeks after irradiation, the end of the pneumonitis phase of damage, to higher doses compared with these data after whole-lung irradiation. In addition, the slopes of the dose-response curves for irradiation of partial lung volumes were more shallow compared to those after whole-lung irradiation. Increases in breathing rate correlated with lethality when the volume irradiated was equal to or greater than 50% of the reference volume. However, after irradiation of volumes smaller than 40%, breathing rate increases were not accompanied by death. A heterogeneous response of the mouse lung to radiation was observed in the first experiment and confirmed by the second experiment. For a given volume irradiated, the isoeffect dose was always less for the base than for the apex of the lung. The threshold volume for breathing rate changes was less than 17 and 40% when the irradiated volumes involved the apex and base, respectively. For lethality, the threshold volume was between 40 and 70% for the base and greater than 50% for the apex of the lung. Finally, damage as assessed by histological evidence of pneumonitis was observed in the irradiated area only.

CONCLUSIONS

(a) The volume effect was resolvable in mice, (b) the volume effect in mouse lung exhibits a clear threshold for morbidity, (c) the threshold volume for morbidity is dependent on the end point, (d) the response of mouse lung is heterogeneous, dependent on the site irradiated, and is always greater for the same volumes irradiated in the base than the apex, and, (e) histopathological damage does not always produce observable morbidity.

Dosimetric considerations in treating mediastinal disease with mantle fields: characterization of the dose under mantle blocks

PURPOSE

While the rationale for using mantle fields is well understood and the prescription of these fields is straightforward, the underlying complexity of the dose distributions that result is not generally appreciated. This is especially true in the choice of lung block design, which affects the dose to both the target volume as well as to the normal lung tissue. The key to the design of optimal lung blocks is the physician's perception of the complex relationship between the geometric and dosimetric aspects of heavily modified fields, as well as how the physical and anatomical properties of the target volume and the shape of the patient's lungs relate to the images visualized on simulator films.

METHODS AND MATERIALS

Depth doses and cross-beam profiles of blocks ranging in width from 1 cm to 10 cm were taken using an automated beam scanning system. These data were then converted to "shadow fields." The results were compared to open fields of the same size using standard methodology.

RESULTS

Shadow fields behave quite similarly to small, open fields in terms of x-ray-light field congruence, flatness, symmetry, and penumbra. There is a 2-3 mm rim between the edge of the block and the point at which it becomes nominally effective. The dose at the center of a block, which gives the normalization of the shadow fields, is given by a block transmission factor (BTF), which produces results in excellent agreement with measurements over a wide variety of block sizes and tissue depths.

CONCLUSION

The radiation dose under shielding blocks can be considerably higher than expected, and care must be exercised when drawing blocks close to critical structures. The effects of blocks can be described in terms of normalized shadow fields, which behave similar to narrow, open fields, but with a divergence characteristic of their position relative to the radiation source. The normalization value for these fields, which gives the relative dose under the block, can be obtained from a straightforward analytical expression, the BTF.

Simulation accuracy in radiotherapy for maxillary sinus tumors

PURPOSE

To evaluate the accuracy and clinical importance of beam positioning during simulation of radiation treatment for tumors in the maxillary sinus.

METHODS AND MATERIALS

Five patients were prepared as if they were to be treated for a maxillary sinus tumor. A three-beam computed tomography (CT) scan-based computer plan was made for each patient. The location of the central beam axis of each beam was measured, relative to bony anatomical structures. A simulation was performed using the bony references to position the radiation beams during simulation. After this, the simulation procedure was repeated by the use of a noninvasive external localization frame with a known accuracy and reproducibility within 2 mm margins.

RESULTS

When defining the clinical target volume as the known tumor with a 1 cm margin, three out of five patients would suffer a partial geographical miss throughout the entire radiation treatment due to erroneous beam positioning at the simulation stage when using bony structures as a guide for beam positioning. The influence of these errors is analyzed as normal tissue complication and tumor control probabilities.

CONCLUSION

When defining a planning target volume, one should consider a margin to correct for possible simulation errors. We advise the use of objective, external (and thus nonanatomical) landmarks as a reference during simulation to reduce this extra margin to a minimum. In case of simulation, using bony structures as a reference, an additional margin should be entered, depending on the simulation accuracy that can be obtained.

A method for implementing dynamic photon beam intensity modulation using independent jaws and a multileaf collimator

A mathematical model is derived for digitally controlled linear accelerators to deliver a desired photon intensity distribution by combining collimator motion and machine dose rate variations. It shows that, at any instant, the quotient of the machine dose rate and the speed of collimator motion is proportional to the gradient of the desired in-air photon fluence distribution. The model is applicable for both independently controlled collimator jaws and multileaf collimators and can be implemented by controlling different parameters to accommodate linear accelerators from different manufactures. For independent jaws, each pair of jaws creates photon fluence variations along the direction of the jaw movement. For multileaf collimators, where each leaf is independently controlled, any two-dimensional (2D) photon fluence distribution can be delivered. The model has been implemented for wedged isodose distributions using independent jaws, and 2D intensity modulation using a multileaf collimator. One-dimensional (1D) wedged isodose distributions are created by moving an independent jaw at constant speed while varying machine dose rate. 2D intensity modulation has been implemented using a 'dynamic stepping' scheme, which controls the leaf progression during irradiation at constant machine dose rate. With this automated delivery scheme, the beam delivery time for dynamic intensity modulation, which depends on the complexity of the desired intensity distribution, approaches that of conventional beam modifiers. This paper shows the derivation of the model, its application, and our delivery scheme. Examples of 1D dynamic wedges and 2D intensity modulations will be given to illustrate the versatility of the model, the simplicity of its application, and the efficiency of beam delivery. These features make this approach practical for delivering conformal therapy treatments.

Analysis of clinical complication data for radiation hepatitis using a parallel architecture model

PURPOSE

The detailed knowledge of dose volume distributions available from the three-dimensional (3D) conformal radiation treatment of tumors in the liver (reported elsewhere) offers new opportunities to quantify the effect of volume on the probability of producing radiation hepatitis. We aim to test a new parallel architecture model of normal tissue complication probability (NTCP) with these data.

METHODS AND MATERIALS

Complication data and dose volume histograms from a total of 93 patients with normal liver function, treated on a prospective protocol with 3D conformal radiation therapy and intraarterial hepatic fluorodeoxyuridine, were analyzed with a new parallel architecture model. Patient treatment fell into six categories differing in doses delivered and volumes irradiated. By modeling the radiosensitivity of liver subunits, we are able to use dose volume histograms to calculate the fraction of the liver damaged in each patient. A complication results if this fraction exceeds the patient's functional reserve. To determine the patient distribution of functional reserves and the subunit radiosensitivity, the maximum likelihood method was used to fit the observed complication data.

RESULTS

The parallel model fit the complication data well, although uncertainties on the functional reserve distribution and subunit radiosensitivity are highly correlated.

CONCLUSION

The observed radiation hepatitis complications show a threshold effect that can be described well with a parallel architecture model. However, additional independent studies are required to better determine the parameters defining the functional reserve distribution and subunit radiosensitivity.

External beam radiotherapy of localized prostatic adenocarcinoma. Evaluation of conformal therapy, field number and target margins

The purpose of the present study was to identify factors of importance in the planning of external beam radiotherapy of prostatic adenocarcinoma. Seven patients with urogenital cancers were planned for external radiotherapy of the prostate. Four different techniques were used, viz. a 4-field box technique and four-, five- or six-field conformal therapy set-ups combined with three different margins (1-3 cm). The evaluations were based on the doses delivered to the rectum and the urinary bladder. A normal tissue complication probability (NTCP) was calculated for each plan using Lyman's dose volume reduction method. The most important factors that resulted in a decrease of the dose delivered to the rectum and the bladder were the use of conformal therapy and smaller margins. Conformal therapy seemed more important for the dose distribution in the urinary bladder. Five- and six-field set-ups were not significantly better than those with four fields. NTCP calculations were in accordance with the evaluation of the dose volume histograms. To conclude, four-field conformal therapy utilizing reduced margins improves the dose distribution to the rectum and the urinary bladder in the radiotherapy of prostatic adenocarcinoma.

The tetrad and hexad: maximum beam separation as a starting point for noncoplanar 3D treatment planning: prostate cancer as a test case

PURPOSE

In contrast to computer optimized three-dimensional (3D) treatment planning, we have used maximally separated, noncoplanar beams as the starting point for 3D treatment planning of prostate cancer to maximize the rate of dose fall off from the target volume and minimize dose to surrounding tissues.

MATERIALS AND METHODS

A planar four-field plan, a planar six-field plan, a tetrad plan, and a hexad plan are analyzed using a 3D treatment planning system which is capable of displaying real-time 3D dose distributions within volume reconstructed data sets (VISTAnet--an extension of the virtual simulator). The tetrad plan is based on the methane molecule and the hexad plan has a minimum separation of 58 degrees on beam entrance. All fields are conformal. The irradiated volume equals the clinical target volume plus a 1 cm margin. Competing plans are compared using cumulative dose-volume histograms and normal tissue complication probabilities.

RESULTS

The crossover point, the isodose surface that conforms more to the beams than the target, is introduced and described. The hexad and tetrad plans result in tighter dose distributions when compared to the planar plans with the same number of beams. The tetrad plan treats a volume less than or equal to the planar six-field plan at isodose surfaces above 18% except between 37% and 44% where the tetrad volume is slightly larger. As expected from integral dose considerations, the amount of normal tissue receiving some radiation increases, but the amount receiving clinically significant amounts of radiation decreases as the number of beams increase. The plan involving the largest number of noncoplanar beams results in the tightest isodose distribution. Analysis of rectal and bladder cumulative dose volume histograms does not reveal a clearly superior plan based on normal tissue complication probabilities.

CONCLUSIONS

Using basic principles of solid geometry, maximally separated beams without significant overlap on exit or entrance can be designed which minimize clinically significant dose to surrounding tissues and tighten the isodose distribution around the target volume. The emphasis of this treatment plan optimization is geometric in contrast to methods using computer optimization or artificial intelligence.

The potential and limitations of the inverse radiotherapy technique

The objective of the work presented in this paper is to explore the scope of the applicability of the inverse radiotherapy technique for designing optimized intensity distributions to achieve a desired dose distribution. A specified desired uniform dose to the target volume is inverted, subject to constraints on the surrounding normal tissue dose, to produce optimum intensity distributions in a set of beams arranged around the target volume. We employed the inverse technique and software developed by Bortfeld and evaluated results both qualitatively and quantitatively using dose distribution displays, dose-volume histograms and biological indices including tumor control probability and normal tissue complication probabilities. So far we have applied this methodology to prostate and lung treatment plans. For prostate the inverse technique produces satisfactory approximations of the desired dose distributions. However, for lung its performance is considerably inferior. Our investigations point to a number of factors for this difference, the primary ones being differences in the tolerance doses of neighboring normal tissues, magnitudes of volume effect, tissue architectures, and the achievability of the specified desired dose distributions. We conclude that, for certain clinical situations, it is not sufficient to specify the objectives of optimization purely in terms of the desired pattern of the dose. The objectives must also include dose-volume effects and biological indices. Furthermore, the mathematics of optimization must be able to incorporate these factors into the process. We find that the inverse technique is not suitable for situations where dose-volume considerations and biological indices are important and that other methods of optimization of intensity distributions should be explored.

Expanding the use and effectiveness of dose-volume histograms for 3-D treatment planning. I: Integration of 3-D dose-display

PURPOSE

A technique is presented for overcoming a major deficiency of histogram analysis in three-dimensional (3-D) radiotherapy treatment planning; the lack of spatial information.

METHODS AND MATERIALS

In this technique, histogram data and anatomic images are displayed in a side-by-side fashion. The histogram curve is used as a guide to interactively probe the nature of the corresponding 3-D dose distribution. Regions of dose that contribute to a specific dose bin or range of bins are interactively highlighted on the anatomic display as a window-style cursor is positioned along the dose-axis of the histogram display. This dose range highlighting can be applied to two-dimensional (2-D) images and to 3-D views which contain anatomic surfaces, multimodality image data, and representations of radiation beams and beam modifiers. Additionally, as a range of histogram bins is specified, dose and volume statistics for the range are continually updated and displayed.

RESULTS

The implementation of these techniques is presented and their use illustrated for a nonaxial three field treatment of a hepatic tumor.

CONCLUSION

By integrating displays of 3-D doses and the corresponding histogram data, it is possible to recover the positional information inherently lost in the calculation of a histogram. Important questions such as the size and location of hot spots in normal tissues and cold spots within target volumes can be more easily uncovered, making the iterative improvement of treatment plans more efficient.

Volume effects in rhesus monkey spinal cord

PURPOSE

An experiment was conducted to test for the existence of a volume effect in radiation myelopathy using Rhesus monkeys treated with clinically relevant field sizes and fractionation schedules.

METHODS AND MATERIALS

Five groups of Rhesus monkeys were irradiated using 2.2 Gy per fraction to their spinal cords. Three groups were irradiated with 8 cm fields to total doses of 70.4, 77, and 83.6 Gy. Two additional groups were irradiated to 70.4 Gy using 4 and 16 cm fields. The incidence of paresis expressed within 2 years following the completion of treatment was determined for each group. Maximum likelihood estimation was used to determine parameters of a logistic dose response function. The volume effect was modeled using the probability model in which the probability of producing a lesion in an irradiated volume is governed by the probability of the occurrence of independent events. This is a two parameter model requiring only the estimates of the parameters of the dose-response function for the reference volume, but not needing any additional parameters for describing the volume effect.

RESULTS

The probability model using a logistic dose-response function fits the data well with the D50 = 75.8 Gy for the 8-cm field. No evidence was seen for a difference in sensitivities for different anatomical levels of the spinal cord. Most lesions were type 3, combined white matter parenchymal and vascular lesions. Latent periods did not differ significantly from those of type 3 lesions in humans.

CONCLUSION

The spinal cord exhibits a volume effect that is well described by the probability model. Because the dose response function for radiation myelopathy is steep, the volume effect is modest. The Rhesus monkey remains the animal model most similar to humans in dose response, histopathology, and latency for radiation myelopathy.

Dose-volume histogram and 3-D treatment planning evaluation of patients with pneumonitis

PURPOSE

Tolerance of normal lung to inhomogeneous irradiation of partial volumes is not well understood. This retrospective study analyzes three-dimensional (3-D) dose distributions and dose-volume histograms for 63 patients who have had normal lung irradiated in two types of treatment situations.

METHODS AND MATERIALS

3-D treatment plans were examined for 21 patients with Hodgkin's disease and 42 patients with nonsmall-cell lung cancer. All patients were treated with conventional fractionation, with a dose of 67 Gy (corrected) or higher for the lung cancer patients. A normal tissue complication probability description and a dose-volume histogram reduction scheme were used to assess the data. Mean dose to lung was also calculated.

RESULTS

Five Hodgkin's disease patients and nine lung cancer patients developed pneumonitis. Data were analyzed for each individual independent lung and for the total lung tissue (lung as a paired organ). Comparisons of averages of mean lung dose and normal tissue complication probabilities show a difference between patients with and without complications. Averages of calculated normal tissue complication probabilities for groups of patients show that empirical model parameters correlate with actual complication rates for the Hodgkin's patients, but not as well for the individual lungs of the lung cancer patients treated to larger volumes of normal lung and high doses.

CONCLUSION

This retrospective study of the 3-D dose distributions for normal lung for two types of treatment situations for patients with irradiated normal lung gives useful data for the characterization of the dose-volume relationship and the development of pneumonitis. These data can be used to help set up a dose escalation protocol for the treatment of nonsmall-cell lung cancer.

A comparison of three inverse treatment planning algorithms

Three published inverse treatment planning algorithms for physical optimization of external beam radiotherapy are compared. All three algorithms attempt to minimize a quadratic objective function of the dose distribution. It is shown that the algorithms are based on the common framework of Newton's method of multi-dimensional function minimization. The approximations used within this framework to obtain the different algorithms are described. The use of these algorithms requires that the number of weights of elemental dose distributions be equal to the number of sample points taken in the dose volume. The primary factor in determining how the algorithms are implemented is the dose computation model. Two of the algorithms use pencil beam dose models and therefore directly optimize individual pencil beam weights, whereas the third algorithm is implemented to optimize groups of pencil beams, each group converging upon a common point. All dose computation models assume that the irradiated medium is homogeneous. It is shown that the two different implementations produce similar results for the simple optimization problem of conforming dose to a convex target shape. Complex optimization problems consisting of non-convex target shapes and dose limiting structures are shown to require a pencil beam optimization method.

Advances in 3-dimensional radiation treatment planning systems: room-view display with real time interactivity

PURPOSE

We describe our 3-dimensional (3-D) radiation treatment planning system for external photon and electron beam 3-D treatment planning which provides high performance computational speed and a real-time display which we have named "room-view" in which the simulated target volumes, critical structures, skin surfaces, radiation beams and/or dose surfaces can be viewed on the display monitor from any arbitrary viewing position.

METHODS AND MATERIALS

We have implemented the 3-D planning system on a graphics superworkstation with parallel processing. Patient's anatomical features are extracted from contiguous computed tomography scan images and are displayed as wireloops or solid surfaces. Radiation beams are displayed as a set of diverging rays plus the polygons formed by the intersection of these rays with planes perpendicular to the beam axis. Controls are provided for each treatment machine motion function. Photon dose calculations are performed using an effective pathlength algorithm modified to accommodate 3-D off-center ratios. Electron dose calculations are performed using a 3-D pencil beam model.

RESULTS

Dose distribution information can be displayed as 3-D dose surfaces, dose-volume histograms, or as isodoses superimposed on 2-D gray scale images of the patient's anatomy. Tumor-control-probabilities, normal-tissue-complication probabilities and a figure-of-merit score function are generated to aid in plan evaluation. A split-screen display provides a beam's-eye-view for beam positioning and design of patient shielding block apertures and a concurrent "room-view" display of the patient and beam icon for viewing multiple beam set-ups, beam positioning, and plan evaluation. Both views are simultaneously interactive.

CONCLUSION

The development of an interactive 3-D radiation treatment planning system with a real-time room-view display has been accomplished. The concurrent real-time beam's-eye-view and room-view display significantly improves the efficacy of the 3-D planning process.

Use of Veff and iso-NTCP in the implementation of dose escalation protocols

PURPOSE

This report investigates the use of a normal tissue complication probability (NTCP) model, 3-D dose distributions, and a dose volume histogram reduction scheme in the design and implementation of dose escalation protocols for irradiation of sites that are primarily limited by the dose to a normal tissue which exhibits a strong volume effect (e.g., lung, liver).

METHODS AND MATERIALS

Plots containing iso-NTCP contours are generated as a function of dose and partial volume using a parameterization of a NTCP description. Single step dose volume histograms are generated from 3-D dose distributions using the effective-volume (Veff) reduction scheme. In this scheme, the value of Veff for each dose volume histogram is independent of dose units (Gy, %). Thus, relative dose distributions (%) may be used to segregate patients by Veff into bins containing different ranges of Veff values before the assignment of prescription doses (Gy). The doses for each bin of Veff values can then be independently escalated between estimated complication levels (iso-NTCP contours).

RESULTS AND CONCLUSION

Given that for the site under study, an investigator believes that the NTCP parameterization and the Veff methodology at least describe the general trend of clinical expectations, the concepts discussed allow the use of patient specific 3-D dose/volume information in the design and implementation of dose escalation studies. The result is a scheme with which useful prospective tolerance data may be systematically obtained for testing the different NTCP parameterizations and models.

Renal tolerance to nonhomogenous irradiation: comparison of observed effects to predictions of normal tissue complication probability from different biophysical models

PURPOSE

A patient series was analyzed retrospectively as an example of whole organ kidney irradiation with an inhomogenous dose distribution to test the validity of biophysical models predicting normal tissue tolerance to radiotherapy.

METHODS AND MATERIAL

From 1969 to 1984, 142 patients with seminoma were irradiated to the paraaortic region using predominantly rotational techniques which led to variable but partly substantial exposure of the kidneys. Median follow up was 8.2 (2.1-21) years and actuarial 10-year survival (Kaplan-Meier estimate) 82.8%. For all patients 3-dimensional dose distributions were reconstructed and normal tissue complication probabilities for the kidneys were generated from the individual dose volume histograms. To this respect different published biophysical algorithms had been introduced in a 3-dimensional-treatment planning system.

RESULTS

In seven patients clinical manifest renal impairment was observed (interval 10-84 months). An excellent agreement between predicted and observed effects was seen for two volume-oriented models, whereas complications were overestimated by an algorithm based on critical element assumptions.

CONCLUSIONS

Should these observations be confirmed and extended to different types of organs corresponding algorithms could easily be integrated into 3-dimensional-treatment planning programs and be used for comparing and judging different plans on a more biologically oriented basis.

Optimization by simulated annealing of three-dimensional, conformal treatment planning for radiation fields defined by a multileaf collimator: II. Inclusion of two-dimensional modulation of the x-ray intensity

Interest is rapidly growing in using multiple x-radiation fields defined by a multileaf collimator to achieve conformal radiotherapy. Three-dimensional treatment planning in such situations is in its infancy and most 3D planning systems provide no tools for optimizing therapy. A previous paper addressed how to calculate optimum beamweights when both the target volume and all or some parts of organs at risk were in the fields-of-view. This work allowed a maximum of two weights per field. The present paper extends this technique to allow each radiation port to be spatially modulated across the geometrically shaped field. An optimization method based on simulated annealing is presented. It is shown that including spatial modulation leads to a wider separation between the dose-volume histograms of the target volume and organs at risk. The improvement is quantified in terms of the tumour control probability at constant normal tissue complication probability. Possible limitations of the a posteriori applied biological model are discussed in detail.

Three-dimensional photon treatment planning for Hodgkin's disease

A multi-institutional study was undertaken using computerized planning systems to develop three-dimensional (3-D) radiotherapy plans for Hodgkin's disease (H.D.). Two patients, the first afflicted with bulky stage II disease and another one with early stage I H.D., were studied. Three main categories of plan were produced for each patient: a) a traditional plan which modelled a conventional mantle treatment on the 3-D system, b) a 3-D standard plan where anterior and posterior fields were designed to cover 3-D target volumes, and c) a 3-D unconstrained plan where innovational techniques were employed. Three-dimensional planning provides information about the dose distribution throughout the large volume irradiated in patients with H.D. that is not available with conventional mantle planning. The use of 3-D techniques resulted in improved tumor coverage, but by allowing for uncertainties such as motion, the doses to normal tissues tended to be higher. The use of unorthodox beam arrangements introduced added complexities, and further increased the lung doses. The most even dose distributions were obtained by incorporating compensating filters into anterior fields. Clinicians showed wide variations in their assessment of the plans, possible reasons for which are addressed in this paper. In addition, calculated probabilities from models of tumor control and normal tissue damage are also presented.

Three-dimensional treatment planning for lung cancer

The experience of four institutions involved in a three-dimensional treatment planning contract (NCI) for lung cancer is described. It was found that three-dimensional treatment planning has a significant potential for optimization of treatment plans for radiotherapy of lung cancer both for tumor coverage and sparing of critical normal tissues within the complex anatomy of the human thorax. Evaluation tools, such as dose-volume histograms, and three-dimensional isodose displays, such as multiple plane views, surface dose displays, etc., were found to be extremely valuable in evaluation and comparison of these complex plans. It is anticipated that with further developments in three-dimensional simulation and treatment delivery systems, major progress towards uncomplicated local regional control of lung cancer may be forthcoming.

Fitting of normal tissue tolerance data to an analytic function

During external beam radiotherapy, normal tissues are irradiated along with the tumor. Radiation therapists try to minimize the dose of normal tissues while delivering a high dose to the target volume. Often this is difficult and complications arise due to irradiation of normal tissues. These complications depend not only on the dose but also on volume of the organ irradiated. Lyman has suggested a four-parameter empirical model which can be used to represent normal tissue response under conditions of uniform irradiation to whole and partial volumes as a function of the dose and volume irradiated. In this paper, Lyman's model has been applied to a compilation of clinical tolerance data developed by Emami et al. The four parameters to characterize the tissue response have been determined and graphical representations of the derived probability distributions are presented. The model may, therefore, be used to interpolate clinical data to provide estimated normal tissue complication probabilities for any combination of dose and irradiated volume for the normal tissues and end points considered.

Three-dimensional photon treatment planning for carcinoma of the nasopharynx

The role of 3-D treatment planning for carcinoma of the nasopharynx was assessed in a four institution study. Two patients were worked up and had an extensive number of CT scans on which target volumes and normal tissues were defined. Treatment planning was then performed using state of the art dose planning systems for these patients to assess the value of the new technology. In general, it was demonstrated that multi-field conformal plans could achieve good tumor dose coverage, while at the same time reducing normal tissue doses, compared to standard treatment planning techniques. The role of inhomogeneity corrections, beam energy, and the use of CT vs. simulation films for defining target volumes were also discussed. In addition, techniques to evaluate 3-D plans for the nasopharynx were considered, and some analysis of this problem is presented in this paper.

Three-dimensional treatment planning for para-aortic node irradiation in patients with cervical cancer

Three-dimensional treatment planning has been used by four cooperating centers to prepare and analyze multiple treatment plans on two cervix cancer patients. One patient had biopsy-proven and CT-demonstrable metastasis to the para-aortic nodes, while the other was at high risk for metastatic involvement of para-aortic nodes. Volume dose distributions were analyzed, and an attempt was made to define the role of 3-D treatment planning to the para-aortic region, where moderate to high doses (50-66 Gy) are required to sterilize microscopic and gross metastasis. Plans were prepared using the 3-D capabilities for tailoring fields to the target volumes, but using standard field arrangements (3-D standard), and with full utilization of the 3-D capabilities (3-D unconstrained). In some but not all 3-D unconstrained plans, higher doses were delivered to the large nodal volume and to the volume containing gross nodal disease than in plans analyzed but not prepared with full 3-D capability (3-D standard). The small bowel was the major dose limiting organ. Its tolerance would have been exceeded in all plans which prescribed 66 Gy to the gross nodal mass, although some reduction in small bowel near-maximum dose was achieved in the 3-D unconstrained plans. All plans were able to limit doses to other normal organs to tolerance levels or less, with significant reductions seen in doses to spinal cord, kidneys, and large bowel in the 3-D unconstrained plans, as compared to the 3-D standard plans. A high probability of small bowel injury was detected in one of four 3-D standard plans prescribed to receive 50 Gy to the large para-aortic nodal volume; the small bowel dose was reduced to an acceptable level in the corresponding 3-D unconstrained plan. An optimum beam energy for treating this site was not identified, with plans using 4, 6, 10, 15, 18, and 25 MV photons all being equally acceptable. Attempts to deliver moderate or high doses (50-66 Gy) to this region should be made only after careful analysis of the plan with techniques similar to those employed in this study.

The role of uncertainty analysis in treatment planning

The role of uncertainty analysis in 3-D treatment planning systems was addressed by four institutions which contracted with NCI to evaluate high energy photon external beam treatment planning. Treatment plans were developed at eight disease sites and the effects of uncertainties assessed in a number of experiments. Uncertainties which are patient-site specific included variations in the delineation of target volumes and normal tissues and the effects of positional uncertainties due to physiological motion and setup nonreproducibility. These were found to have a potentially major impact on the doses to the target volumes and to critical normal tissues which could result in significantly altered probabilities of tumor control and normal tissue complications. Other uncertainties, such as the conversion of CT data to electron densities, heterogeneities and dose calculation algorithms' weaknesses, are related to physical processes. The latter was noted to have the greatest potential contribution to uncertainty in some sites. A third category of uncertainty related to the treatment machine, the consequences of compensator misregistration, are exclusive to the site and the treatment portal. Because conventional treatment planning systems have not incorporated uncertainty analysis, tools and techniques had to be devised for this work; further development in this area is needed. Many of the analyses could not have been done without full 3-D capabilities of the planning systems, and it can be anticipated that the availability of uncertainty analysis in these systems which allow nontraditional beam arrangements will be of great value.

Numerical scoring of treatment plans

This is a report on numerical scoring techniques developed for the evaluation of treatment plans as part of a four-institution study of the role of 3-D planning in high energy external beam photon therapy. A formal evaluation process was developed in which plans were assessed by a clinician who displayed dose distributions in transverse, sagittal, coronal, and arbitrary oblique planes, viewed dose-volume histograms which summarized dose distributions to target volumes and the normal tissues of interest, and reviewed dose statistics which characterized the volume dose distribution for each plan. In addition, tumor control probabilities were calculated for each biological target volume and normal tissue complication probabilities were calculated for each normal tissue defined in the agreed-upon protocols. To score a plan, the physician assigned a score for each normal tissue to reflect possible complications; for each target volume two separate scores were assigned, one representing the adequacy of tumor coverage, the second the likelihood of a complication. After scoring each target and normal tissue individually, two summary scores were given, one for target coverage, the second reflecting the impact on all normal tissues. Finally, each plan was given an overall rating (which could include a downgrading of the plan if the treatment was judged to be overly complex).