A radiomics model from joint FDG-PET and MRI texture features for the prediction of lung metastases in soft-tissue sarcomas of the extremities.

This study aims at developing a joint FDG-PET and MRI texture-based model for the early evaluation of lung metastasis risk in soft-tissue sarcomas (STSs). We investigate if the creation of new composite textures from the combination of FDG-PET and MR imaging information could better identify aggressive tumours. Towards this goal, a cohort of 51 patients with histologically proven STSs of the extremities was retrospectively evaluated. All patients had pre-treatment FDG-PET and MRI scans comprised of T1-weighted and T2-weighted fat-suppression sequences (T2FS). Nine non-texture features (SUV metrics and shape features) and forty-one texture features were extracted from the tumour region of separate (FDG-PET, T1 and T2FS) and fused (FDG-PET/T1 and FDG-PET/T2FS) scans. Volume fusion of the FDG-PET and MRI scans was implemented using the wavelet transform. The influence of six different extraction parameters on the predictive value of textures was investigated. The incorporation of features into multivariable models was performed using logistic regression. The multivariable modeling strategy involved imbalance-adjusted bootstrap resampling in the following four steps leading to final prediction model construction: (1) feature set reduction; (2) feature selection; (3) prediction performance estimation; and (4) computation of model coefficients. Univariate analysis showed that the isotropic voxel size at which texture features were extracted had the most impact on predictive value. In multivariable analysis, texture features extracted from fused scans significantly outperformed those from separate scans in terms of lung metastases prediction estimates. The best performance was obtained using a combination of four texture features extracted from FDG-PET/T1 and FDG-PET/T2FS scans. This model reached an area under the receiver-operating characteristic curve of 0.984 ± 0.002, a sensitivity of 0.955 ± 0.006, and a specificity of 0.926 ± 0.004 in bootstrapping evaluations. Ultimately, lung metastasis risk assessment at diagnosis of STSs could improve patient outcomes by allowing better treatment adaptation.

Poster - Thur Eve - 31: RapidArc total body photon irradiation: A feasibility study.

Total body photon irradiation (TBI) may be delivered with a number of standard techniques, typically using extended SSD geometries to obtain large field sizes. Since 1982 we have treated over 620 patients (adult and pediatric) mostly in the floor-located prone/supine treatment position with AP/PA beams using a column mounted 4 MV linac, and more recently with a fixed field, extended SSD, Cobalt teletherapy unit. This treatment position has many advantages for TBI including ease of delivery (especially for pediatric or compromised patients), improved dose uniformity, simplicity for partial or complete organ shielding, and imaging capabilities to name a few. In an effort to transfer this technique to a conventional isocentric linac, we have undertaken a feasibility study for RapidArc based extended SSD total body irradiation. Proof of principle was first explored using the DBD (Dynamic Beam Delivery) toolbox to configure a 6 MV beam with an 80 degree arc, centered on the 12 o'clock gantry angle with a sliding-window beam. This was followed by the development of an ECLIPSE generated 80 degree RapidArc plan. Initial measurements were conducted with a Varian 21EX using the 6 MV DBD beam to explore characteristics such as PDD, surface dose, off-axis ratios, output, dose per MU, and linearity. Subsequently ECLIPSE generated RapidArc TBI plans using similar partial arcs were also evaluated. Encouraged by our results, we believe this technique shows potential for making floor-located AP/PA total body photon irradiation possible for any standard RapidArc enabled isocentric linac.

SU-D-BRB-03: Prediction of Tumor Outcomes Through Wavelet Image Fusion and Texture Analysis of PET/MR Imaging.

PURPOSE: To investigate the combination of PET/MR image features for the early prediction of tumor metastases to the lungs in soft-tissue sarcoma (STS) cancer. METHODS: A dataset of 24 patients with histologically proven STS was used in this study. All patients underwent pre-treatment FDG-PET and MR scans, which comprised of T1 and T2-fat suppression weighted (T2FS) sequences. The patients had a median follow-up period of 36 months (range: 6-69 months). Eight patients developed metastases to the lungs.Tumors were contoured on the T2FS scans by an expert physician. Fusion of the co-registered FDG-PET/MR scans was performed using a wavelet transform technique. A SUV feature (SUVmax) from the FDG-PET scans and 6 texture features from the co-occurrence matrix of the fused scans were extracted from the tumor region and correlation with the clinical endpoint of metastases to the lungs was investigated. Statistical analysis was performed using Spearman's rank correlation (rs) and multivariable logistic regression. RESULTS: The highest univariate prediction was found on FDG-PET/T2FS fused scans analyzed using the texture features "Sum-Mean" and "Variance". These two fused scan-texture feature combinations reached rs = -0.6838 (p = 0.0003). In comparison, SUVmax reached rs = -0.6257 (p = 0.0011). The highest multivariate prediction was found with the following 3- parameter model: -3.15*SUVmax - 5.37*FDG-PET/T2FS-Sum-Mean + 0.57*FDG-PET/T1-Variance. This model reached rs = 0.7977 (p = 0.000005). CONCLUSIONS: This work indicates the potential of PET/MR texture features of tumors as complementary metrics to existing prognostic factors. Substantial improvement in terms of prediction of metastases to the lungs in STS cancer was found with the combination of texture features from fused FDG-PET/MR scans. Potentially, this could improve patients' outcomes by allowing better adaptation of treatments. Future work will involve evaluation of the robustness of the proposed method and validation on a larger set of patients.

Treatment-related morbidity in atypical teratoid/rhabdoid tumor: multifocal necrotizing leukoencephalopathy.

BACKGROUND: Atypical teratoid/rhabdoid tumor (AT/RT) is an aggressive malignant brain tumor that, since it was first identified, has been treated with aggressive treatment regimens, e.g. high-dose chemotherapy with stem cell rescue and early radiotherapy. We reviewed our experience because of concerns with respect to treatment-related toxicity in our patients. METHODS: Seven patients with a median age at presentation of 18 months were diagnosed with AT/RT between 1996 and 2006. Tumor location was supratentorial in 2 patients, in the posterior fossa in 4 and spinal in 1. Gross total resection was performed in 1 patient, subtotal resection in 5 and biopsy only in 1. Adjuvant treatment consisted of chemotherapy and radiotherapy in 5 patients. RESULTS: Median progression-free survival was 4 months, and median overall survival was 7 months. Two children are alive at 44 and 102 months. Significant surgical and chemotherapy-related morbidity was seen. Biopsy-proven multifocal necrotizing leukoencephalopathy (MNL) was seen in one patient who is alive 44 months after diagnosis. Another patient who was thought to have recurrent tumor in the brainstem 9 months after diagnosis had imaging findings compatible with MNL. CONCLUSION: Although improving results are reported for AT/RT using intensive treatment regimens, treatment-related morbidity is considerable in this young patient population.

The challenge of scoring radiation-induced lung toxicity.

AIMS: The increasing use of curative radiation treatment in lung cancer mandates accurate assessment of late lung toxicity. The Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer (RTOG/EORTC) scoring schema combines clinical symptoms and radiological changes and may be confusing. Some have used a scoring scale modified from the National Cancer Institute Common Toxicity Criteria (NCI-CTC) version 2 scale based on symptoms only. Clinical data using these two different scales have been compared as if they give similar results. The present study compared the outcomes using the two scales in the same group of patients. MATERIALS AND METHODS: The medical records and imaging of patients with non-small cell lung cancer who received definitive radiotherapy were reviewed. Eligible patients had a minimum follow-up of 12 months with no clinical signs of local relapse. Radiation-induced lung toxicity was scored using the RTOG/EORTC and the NCI-CTC scales. RESULTS: In total, 50 patients were analysed. All patients developed radiographic abnormalities after curative radiotherapy. Grade 0, 1, 2 and 3 toxicity was 0, 28, 49 and 23%, respectively, according to the RTOG/EORTC scale and 86, 7, 7 and 0%, respectively, according to the NCI-CTC scale, showing that the inclusion of radiographic abnormalities changes and significantly upgrades the toxicity scores. CONCLUSION: After curative radiotherapy, all patients presented some radiographic abnormality. There was no correlation with lung symptoms. The assessment of radiation-induced lung toxicity differs depending on the scoring system used. Comparison of reports that use different scoring scales should be made with caution. A scale based on symptoms only, such as the NCI-CTC scale, may be more appropriate to evaluate long-term toxicity after curative radiotherapy for lung cancer.

Does adjuvant alpha-interferon improve outcome when combined with total skin irradiation for mycosis fungoides?

BACKGROUND: Patients with mycosis fungoides (MF) experience frequent disease recurrences following total skin electron irradiation (TSEI) and may benefit from adjuvant therapy. OBJECTIVES: To review the McGill experience with adjuvant alpha-interferon (IFN) in the treatment of MF. METHODS: From 1990 to 2000, 50 patients with MF were treated with TSEI: 31 with TSEI alone and 19 with TSEI + IFN. Median TSEI dose was 35 Gy. In the TSEI + IFN group, IFN was given subcutaneously at 3 x 10(6) units three times per week starting 2 weeks prior to start of TSEI, continued concurrently with the radiation and for an additional 12 months following TSEI. The TSEI alone group included 16 men and 15 women with a median age of 61 years (range 31-84). The TSEI + IFN group included 14 men and five women with a median age of 51 years (range 24-83). Clinical stage was IA, IB, IIA, IIB, III and IVA in 2, 9, 4, 8, 1 and 7 patients of the TSEI group and 0, 3, 3, 7, 4 and 2 patients of the TSEI + IFN group. RESULTS: Median follow up for living patients was 70 months. All patients responded to treatment. Complete response (CR) rate was 65% following TSEI and 58% following TSEI + IFN (P = 0.6). Median overall survival (OS) was 61 months following TSEI and 38 months following TSEI + IFN (P = 0.4). Acute grade II-III dermatitis was seen in all patients. Fever, chills or myalgia were seen in 32% of patients treated with TSEI + IFN. CONCLUSIONS: Concurrent IFN and TSEI is feasible, with acceptable toxicity. Even when controlling for disease stage, the addition of IFN did not appear to increase CR rate, disease-free survival or OS.

Effect of delay in initiating radiotherapy for patients with early stage breast cancer.

AIMS: For early stage breast cancer, a standard treatment option is partial mastectomy followed by radiation treatment. The 5-year risk of local recurrence ranges from 6-9%. Variable waiting times for radiation treatment of breast cancer in our institution provided an opportunity to evaluate the impact of waiting time on the risk of local recurrence. MATERIALS AND METHODS: Between January 1988 and December 1989, 482 patients with stage I and II breast cancer were treated with radiotherapy in our institution. Information on prognostic factors, such as age, tumour size, histological grade, number of positive lymph nodes and margins of resection, was abstracted from their charts. The interval between date of surgery and date of initial radiation treatment was noted. Dates of local recurrence, metastasis and deaths were recorded. RESULTS: At 5 years, the local recurrence rate was 8%, the metastatic rate 12% and the 'cause-specific' survival rate 90%. In univariate analysis, the risk of local recurrence was associated with younger age, higher histological grade and time to radiation treatment. In the multivariate analysis, the effect of time to radiation treatment on the risk of local recurrence was equivocal. CONCLUSION: Delay in radiation treatment may be associated with an increased risk of local recurrence of breast cancer, at least in our centre. Future research is needed on a larger data set to more accurately estimate the effect of time to radiation treatment on the risk of local recurrence.

Delay in adjuvant radiation treatment and outcomes of breast cancer--a review.

Recent meta-analyses have shown the importance of locoregional control as a long-term determinant of breast cancer survival. Whether factors related to the delivery of radiotherapy, such as delay, dose, fractionation or irradiated volume, are associated with outcome remains unclear. We performed a critical review of the literature on delay to radiation using a computerized search of papers published between 1985 and 2000. Periods of accrual, details of radiotherapy, surgical and systemic treatment, and information on prognostic factors were noted. Studies on sequencing of adjuvant therapy were compared to studies on delay to radiation, classified according to whether or not patients also received chemotherapy. Comparisons of patients receiving systemic therapy to individuals spared this option were considered uninformative since the impact of delaying radiation is then highly confounded by systemic treatment received. The single published experimental study on sequencing suggests that delay to radiation may compromise local control, and this is consistent with a few retrospective reports on delay to radiotherapy among patients receiving chemotherapy. However, indirect evidence from two randomized clinical trials of chemotherapy, and the majority of observational studies on delay to radiotherapy, suggest that it has no impact on either local, distant control or survival. Factors, methodological, and others, that could explain these inconsistencies are discussed. No study restricted to patients at low risk of recurrence suggested an impact of delaying radiation. Short chemotherapy regimens are likely to represent a safe option with respect to outcome of radiation treatment.

Comparison of manual and computer-generated customized blocks for whole brain fields used in the treatment of medulloblastoma.

BACKGROUND: By using CT simulation and three-dimensional planning, we have developed a simple but accurate method of producing customized blocks for the cranial fields in the treatment of medulloblastoma. We compare here the margins and the volume of normal tissues included in the treatment volume with those obtained using blocks generated manually. PROCEDURE: Customized blocks for the whole brain field are generated using CT planning and autoblock function. The clinical target volume (CTV) is defined as the whole cerebrospinal fluid pathway, and the whole brain and spinal cord are contoured. A margin of 1.1 cm is generated using the autoblock function to account for set-up errors (3-5 mm) and penumbra (approximately 7 mm). A separate set of blocks was generated manually without the knowledge of the ones generated by the CT-simulator. These 2 sets of blocks were compared for a cohort of 7 consecutive patients. RESULTS: Overall, the manual blocks and the computer-generated blocks were quite similar. Those generated manually had more variations; they were always tighter (median of 6 mm tighter; range: 3-7 mm) at the level of the cribriform plate and in 5/7 patients were more generous (median of 6 mm more generous, range: 0-6 mm) at the temporal lobes. Dosimetric analysis showed that both methods provide adequate coverage of the CTV, with 100% of the CTV receiving > 95% of the prescribed dose for both. CONCLUSIONS: The customized block method for whole brain fields is simple to use and ensures adequate coverage of the target volume.

Brainstem Gliomas. A 10-year institutional review.

Case records of 37 patients with a diagnosis of brainstem glioma treated at the Montreal Children's Hospital from June 1989 to June 1999 were reviewed. 15 patients had diffuse pontine gliomas and 22 patients had focal forms of brainstem gliomas. The two groups were compared with respect to age, clinical evolution, radiological appearance, type of surgery practised, histological diagnosis, adjuvant treatments and survival. A non-pontine brainstem location, a cystic or exophytic component, bright enhancement with gadolinium injection, a histological diagnosis of pilocytic astrocytoma or ganglioglioma were favourable prognostic factors. Progression-free survival and overall survival were significantly worse in the group of patients with diffuse pontine gliomas. The relative impact of radical surgery and/or radiotherapy is analysed. Surgery coupled to adjuncts such as navigation, ultrasound and monitoring plays an important role for focal brainstem lesions. Focal/conformal radiotherapy has an adjuvant role but better treatments are needed for the diffuse pontine brainstem lesions.

A detrimental effect of a combined chemotherapy-radiotherapy approach in children with diffuse intrinsic brain stem gliomas?

PURPOSE: To compare the proportion of patients that survive at least 1 year following treatment with hyper-fractionated radiotherapy (HRT) to a dose of 70.2 Gy on Pediatric Oncology Group (POG) study #8495 with that of patients treated with similar radiotherapy plus cisplatinum given by continuous infusion on weeks 1, 3, and 5 of radiotherapy on POG #9239. METHODS AND MATERIALS: The eligibility criteria for the two studies were identical and included age 3 to 21 years, previously untreated tumor involving the brain stem of which two-thirds was in the pons, history less than 6 months, and clinical findings typical for diffuse intrinsic brain stem glioma, including cranial nerve deficits, long tract signs, and ataxia. The outcome of 57 patients who were treated at the 70.2 Gy dose level of POG #8495 between May 1986 and February 1988 was compared with that of 64 patients treated with identical radiotherapy plus cisplatinum on POG #9239 between June 1992 and March 1996. RESULTS: The number of patients accrued to POG #9239 was determined to guarantee that the probability was at least 0.80 of correctly detecting that the 1-year survival rate exceeded that of patients on POG #8495 by 0.2. However, the z value for this test was -1.564, giving a p value of 0.9411. That is, there is almost sufficient evidence to conclude that survival for patients receiving HRT plus cisplatinum on POG #9239 was worse than that for patients receiving the same radiotherapy alone on POG #8495. CONCLUSION: The finding that patients who received cisplatinum given as a radiosensitizing agent concurrent with HRT fared less well than those receiving the same dose of HRT alone was unexpected and is clearly a cause for concern as many current protocols for patients with diffuse intrinsic brain stem gliomas call for use of chemotherapeutic and/or biological agents given concurrent with radiotherapy.

An alternative mantle irradiation technique using 3D CT- based treatment planning for female patients with Hodgkin's disease.

PURPOSE: For female patients, radiotherapy treatment for Hodgkin's disease invariably results in the irradiation of breast tissue that may lead to radiation induced secondary cancers. The risk for secondary breast cancer is correlated with dose. We have developed a technique in an attempt to increase breast sparing during mantle field irradiation for female patients. MATERIAL AND METHODS: To minimize the irradiated breast volume, a virtual simulation technique making use of a Styrofoam breast immobilization board has been developed whereby the patient lies prone with the breasts positioned in grooves within the board. The breast position is adjusted using Styrofoam wedges, and breast placement is verified using an AP CT-pilot view. A CT scan of the neck and thoracic regions is taken, and the lymph nodes, breast volume and critical structures are outlined. Virtual simulation of the mantle fields (typically AP/PA isocentric beams) is performed, and beam blocks are drawn on the digitally reconstructed radiographs (DRR) generated by the virtual simulation package. The shielding is designed to allow adequate margins around the lymph nodes while maximizing shielding of the lung and breast tissues. The para-aortic fields are also easily determined through virtual simulation, where multi-planar reconstructions (MPR) and 3D renderings of the patient's CT data are used to determine the field limits and beam gaps. In addition to allowing for the geometric optimization of the positioning of the breasts under the lung shields, the virtual simulation technique provides the necessary information for a 3D dosimetric analysis, including dose-volume histograms (DVHs) of the irradiated breast volume. RESULTS: The 3D breast sparing technique was qualitatively and quantitatively compared to non-CT-based techniques and other 3D techniques currently available to assess the protection of the breasts. In a preliminary analysis, virtual simulation images (DRRs, 3D rendering and multi-planar reconstruction) demonstrated the advantage of using the breast sparing technique. A further analysis of DVHs showed a reduction of at least 50% in the volume of breast tissue irradiated when using the breast positioning board and virtual simulation as compared to the conventional simulation techniques where a breast immobilization board was not used. CONCLUSIONS: The use of a breast immobilization board and of a virtual simulation technique is recommended for the planning and treatment of female patients with Hodgkin's disease. DVH analysis has shown that this leads to a decrease in the volume of breast irradiated. It is hoped that this approach will reduce the risk of secondary breast malignancies in female patients with Hodgkin's disease.

Designing attenuators for total-body irradiation using virtual simulation.

In total-body photon irradiation, the lungs are the most commonly shielded organ. Lung compensators are often designed by using high-energy portal films. Other organs, such as the kidneys and liver, are poorly visualized in portal films due to their unit-density composition. A computed tomography-based technique to design kidney and liver attenuators involves outlining these organs in a virtual simulation. The position and the shape of the attenuator are then determined from a digitally-reconstructed radiograph. Appropriate attenuator thickness is determined from measured transmission curves. This article provides a summary of this technique for total-body photon irradiation in a 4-MV photon beam.

Stereotactic radiation in primary brain tumors in children and adolescents.

To evaluate treatment outcome and morbidity of stereotactic external-beam irradiation (SEBI) in pediatric patients, we reviewed 14 children treated with SEBI, using a 10-MV isocentric linear accelerator at McGill University between 1988 and 1994. The median follow-up was 46 months (range 6-82 months). The median age was 14 years. There were 8 low-grade astrocytomas, 3 neuromas and 4 other histologies. Twelve patients received fractionated treatments. The median collimator diameter was 2.5 cm (range 1-5 cm). The median biological effective dose delivered to the entire tumor volume was 57 Gy for astrocytomas and 43 Gy for the other histologies. The overall actuarial survival rate and disease-free survival rate at 5 years were 83 and 62%, respectively. For the patients with low-grade astrocytomas, the 5-year survival and disease-free survival rates were 100 and 60%, respectively. Four children had recurrence at a median of 37 months. Four patients developed treatment-related complications: 1 had edema alone, 2 had necrosis and 1 had edema associated with necrosis. Neither the physical nor radiobiological parameters were predictive of the treatment outcome or the treatment complications. Stereotactic irradiation is a valid option for progressive nonresectable tumors in children.

Chemotherapy for brain stem gliomas.

Approximately 80% of tumors arising in the brain stem are diffuse intrinsic lesions. Patients typically present with a short duration of symptoms and signs with significant neurological impairment. Imaging findings are typical, and biopsy is not usually necessary to make a diagnosis. Standard treatment consists of radiotherapy alone. Although the majority of patients will show a significant improvement in neurological status following such treatment, the prognosis is very poor. The median time to disease progression is of the order of 5-6 months, the median survival time less than 1 year, and survival at 2 years and beyond, less than 10%. Over the last 10-15 years, numerous studies have been undertaken in an attempt to develop more effective treatment for children with diffuse intrinsic brain stem tumors. Using a hyperfractionated (twice-daily) schedule, doses of radiotherapy as high as 78 Gy have been given without success. Other approaches use chemotherapy combined with radiotherapy in a variety of different ways, including up-front, concurrent, and adjuvant chemotherapy, and high-dose chemotherapy with rescue. The results of these studies using chemotherapy, as well as other approaches using other systemic agents, are reviewed in detail.

Low-grade astrocytomas in children: evolving management strategies.

Developments in imaging and in neurosurgical techniques over the past decade have substantially altered the management of children with low-grade astrocytoma. Indications for surgery have become more clearly defined, and a larger proportion of children undergo complete or subtotal resection than in the past. Fewer receive adjuvant therapy, even though the options in this regard are more numerous now and theoretically likely to result in less morbidity than conventional external beam radiotherapy. This review will address in particular the correlations between location, imaging appearance, and behavior that need to be more widely appreciated, and present recommendations regarding the management of these tumors.

Pancreatic cancer as a second tumour following treatment of Hodgkin's disease.

Over the past 30 years, the life expectancy in patients with Hodgkin's disease has greatly improved. However, adverse long-term side-effects are now well recognized and development of second malignancies is one of the most important. We report the case of a patient who developed pancreatic cancer 9 years after treatment, with chemotherapy and radiation, for Hodgkin's disease. The increasing number and variety of solid tumours after curative treatment of Hodgkin's disease points to a need for new, less toxic regimens.

Predictors of delay in starting radiation treatment for patients with early stage breast cancer.

PURPOSE: To describe the factors predicting waiting time for radiation treatment in early breast cancer. MATERIALS AND METHODS: Between January 1992 and December 1993, 739 patients with Stage I and II breast cancer were treated with conservative treatment at three McGill University Hospitals. Waiting time was defined as the interval between the date of surgery and the date of the first radiation treatment. Delay was defined as a waiting time of more than 7 weeks for women who did not receive chemotherapy (Group NC, n = 478), and as a waiting time of more than 24 weeks for those who received chemotherapy (Group C, n = 261). We analyzed predictive factors related to the patient (age, stage, treatment on protocol, income by postal code) and to the referring hospital (university or community hospital). RESULTS: For the entire population, 54% of patients were delayed, 72% in Group NC and 21.4% in Group C. Univariate analysis showed an impact of referring hospital in both groups, and of stage and treatment on protocol in Group C (all p = 0.001). Multivariate analysis showed that delays were significantly less in Group NC for women referred from a community hospital (p = 0.001) and in Group C for women with Stage I disease (p = 0.06), those treated on protocol, and those referred from a university hospital (p = 0.001). CONCLUSION: More than half of patients with early breast cancer waited more than the recommended intervals for radiation therapy. However, lower income breast cancer patients did not wait longer for treatment than higher income patients, possibly a result of the Canadian Medicare system which provides universal access to health care.