Albert Murtha

Abbreviated Course of Radiation Therapy in Older Patients With Glioblastoma Multiforme: A Prospective Randomized Clinical Trial

PURPOSE To prospectively compare standard radiation therapy (RT) with an abbreviated course of RT in older patients with glioblastoma multiforme (GBM). PATIENTS AND METHODS One hundred patients with GBM, age 60 years or older, were randomly assigned after surgery to receive either standard RT (60 Gy in 30 fractions over 6 weeks) or a shorter course of RT (40 Gy in 15 fractions over 3 weeks). The primary end point was overall survival. The secondary end points were proportionate survival at 6 months, health-related quality of life (HRQoL), and corticosteroid requirement. HRQoL was assessed using the Karnofsky performance status (KPS) and Functional Assessment of Cancer Therapy-Brain (FACT-Br). RESULTS All patients had died at the time of analysis. Overall survival times measured from randomization were similar at 5.1 months for standard RT versus 5.6 months for the shorter course (log-rank test, P =.57). The survival probabilities at 6 months were also similar at 44.7% for standard RT versus 41.7% for the shorter course (lower-bound 95% CI, -13.7). KPS scores varied markedly but were not significantly different between the two groups (Wilcoxon test, P =.63). Low completion rates of the FACT-Br (45%) precluded meaningful comparisons between the two groups. Of patients completing RT as planned, 49% of patients (standard RT) versus 23% required an increase in posttreatment corticosteroid dosage (chi(2) test, P =.02). CONCLUSION There is no difference in survival between patients receiving standard RT or short-course RT. In view of the similar KPS scores, decreased increment in corticosteroid requirement, and reduced treatment time, the abbreviated course of RT seems to be a reasonable treatment option for older patients with GBM.

Radiation plus Procarbazine, CCNU, and Vincristine in Low-Grade Glioma

BACKGROUND Grade 2 gliomas occur most commonly in young adults and cause progressive neurologic deterioration and premature death. Early results of this trial showed that treatment with procarbazine, lomustine (also called CCNU), and vincristine after radiation therapy at the time of initial diagnosis resulted in longer progression-free survival, but not overall survival, than radiation therapy alone. We now report the long-term results. METHODS We included patients with grade 2 astrocytoma, oligoastrocytoma, or oligodendroglioma who were younger than 40 years of age and had undergone subtotal resection or biopsy or who were 40 years of age or older and had undergone biopsy or resection of any of the tumor. Patients were stratified according to age, histologic findings, Karnofsky performance-status score, and presence or absence of contrast enhancement on preoperative images. Patients were randomly assigned to radiation therapy alone or to radiation therapy followed by six cycles of combination chemotherapy. RESULTS A total of 251 eligible patients were enrolled from 1998 through 2002. The median follow-up was 11.9 years; 55% of the patients died. Patients who received radiation therapy plus chemotherapy had longer median overall survival than did those who received radiation therapy alone (13.3 vs. 7.8 years; hazard ratio for death, 0.59; P=0.003). The rate of progression-free survival at 10 years was 51% in the group that received radiation therapy plus chemotherapy versus 21% in the group that received radiation therapy alone; the corresponding rates of overall survival at 10 years were 60% and 40%. A Cox model identified receipt of radiation therapy plus chemotherapy and histologic findings of oligodendroglioma as favorable prognostic variables for both progression-free and overall survival. CONCLUSIONS In a cohort of patients with grade 2 glioma who were younger than 40 years of age and had undergone subtotal tumor resection or who were 40 years of age or older, progression-free survival and overall survival were longer among those who received combination chemotherapy in addition to radiation therapy than among those who received radiation therapy alone. (Funded by the National Cancer Institute and others; ClinicalTrials.gov number, NCT00003375.).

Segmenting brain tumors with conditional random fields and support vector machines

Markov Random Fields (MRFs) are a popular and well-motivated model for many medical image processing tasks such as segmentation. Discriminative Random Fields (DRFs), a discriminative alternative to the traditionally generative MRFs, allow tractable computation with less restrictive simplifying assumptions, and achieve better performance in many tasks. In this paper, we investigate the tumor segmentation performance of a recent variant of DRF models that takes advantage of the powerful Support Vector Machine (SVM) classification method. Combined with a powerful Magnetic Resonance (MR) preprocessing pipeline and a set of ‘alignment-based’ features, we evaluate the use of SVMs, MRFs, and two types of DRFs as classifiers for three segmentation tasks related to radiation therapy target planning for brain tumors, two of which do not rely on ‘contrast agent’ enhancement. Our results indicate that the SVM-based DRFs offer a significant advantage over the other approaches.

3D Variational Brain Tumor Segmentation using a High Dimensional Feature Set

Tumor segmentation from MRI data is an important but time consuming task performed manually by medical experts. Automating this process is challenging due to the high diversity in appearance of tumor tissue, among different patients and, in many cases, similarity between tumor and normal tissue. One other challenge is how to make use of prior information about the appearance of normal brain. In this paper we propose a variational brain tumor segmentation algorithm that extends current approaches from texture segmentation by using a high dimensional feature set calculated from MRI data and registered atlases. Using manually segmented data we learn a statistical model for tumor and normal tissue. We show that using a conditional model to discriminate between normal and abnormal regions significantly improves the segmentation results compared to traditional generative models. Validation is performed by testing the method on several cancer patient MRI scans.

Quick detection of brain tumors and edemas: A bounding box method using symmetry

A significant medical informatics task is indexing patient databases according to size, location, and other characteristics of brain tumors and edemas, possibly based on magnetic resonance (MR) imagery. This requires segmenting tumors and edemas within images from different MR modalities. To date, automated brain tumor or edema segmentation from MR modalities remains a challenging, computationally intensive task. In this paper, we propose a novel automated, fast, and approximate segmentation technique. The input is a patient study consisting of a set of MR slices, and its output is a subset of the slices that include axis-parallel boxes that circumscribe the tumors. Our approach is based on an unsupervised change detection method that searches for the most dissimilar region (axis-parallel bounding boxes) between the left and the right halves of a brain in an axial view MR slice. This change detection process uses a novel score function based on Bhattacharya coefficient computed with gray level intensity histograms. We prove that this score function admits a very fast (linear in image height and width) search to locate the bounding box. The average dice coefficients for localizing brain tumors and edemas, over ten patient studies, are 0.57 and 0.52, respectively, which significantly exceeds the scores for two other competitive region-based bounding box techniques.

Segmenting Brain Tumors Using Pseudo---Conditional Random Fields

Locating Brain tumor segmentation within MR (magnetic resonance) images is integral to the treatment of brain cancer. This segmentation task requires classifying each voxel as either tumor or nontumor, based on a description of that voxel. Unfortunately, standard classifiers, such as Logistic Regression (LR) and Support Vector Machines (SVM), typically have limited accuracy as they treat voxels as independent and identically distributed (iid). Approaches based on random fields, which are able to incorporate spatial constraints, have recently been applied to brain tumor segmentation with notable performance improvement over iid classifiers. However, previous random field systems involved computationally intractable formulations, which are typically solved using some approximation. Here, we present pseudo-conditional random fields (PCRFs), which achieve accuracy similar to other random fields variants, but are significantly more efficient. We formulate a PCRF as a regularized discriminative classifier that relaxes the classification decision for each voxel by considering the labels and features of neighboring voxels.

Muscle-invasive transitional cell carcinoma of the urinary bladder : A population-based study of patterns of care and prognostic factors

PURPOSE Population-based cancer registries can permit the study of the survivorship of all patients with a particular diagnosis regardless of patterns of referral and practice within a specific geographic distribution. The purpose of this study is to describe the patterns of care, outcome, and prognostic factors for bladder cancer in the northern region of the province of Alberta, Canada, between 1984 and 1993. METHODS AND MATERIALS Between 1984 and 1993, 184 patients from northern Alberta were identified from the Alberta Cancer Registry as having undergone curative treatment for biopsy-proven muscle-invasive transitional cell carcinoma of the bladder. Data were obtained, by retrospective chart review, regarding the staging, pathology, treatment, and outcome of patients treated in the northern Alberta cities of Edmonton, Grande Prairie, and Red Deer, regardless of the responsible treating institution. The prognostic significance of patient-, tumor-, and treatment-related variables were tested using univariate and multivariate analysis using the Cox proportional-hazard model. RESULTS As the primary treatment modality, 74 patients (40%) received radical radiotherapy (RT) without surgery; surgery was used alone in 81 patients (44%), and was combined with preoperative or postoperative radiotherapy in 29 patients (16%). Seventy-three (40%) patients also received concurrent, neoadjuvant, or adjuvant chemotherapy. The Kaplan-Meier estimate of median survival was 2.2 years, and the 5-year overall survival was 30%. Univariate analysis demonstrated the prognostic significance of T classification (p < 0.001), lymph node involvement (p < 0.001), complete response to RT (p = 0.001), hydronephrosis (p = 0.017), and vascular/lymphatic involvement (p = 0.035). Multivariate analysis revealed the following to have a significant association with survival: T classification (p = 0.001), lymph node involvement (p = 0.004), complete response to RT (p = 0.054), hydronephrosis (p = 0.019), and use of chemotherapy in the treatment regimen (p = 0.025). CONCLUSION The strongest prognostic factors in this study were tumor related, and no significant differences in survival were detected between patients treated with primary surgery vs. organ-preservation approaches. A survival advantage associated with the incorporation of chemotherapy into the management schema was detected on multivariate, but not univariate, analysis. Stratification of patients based on tumor characteristics is imperative in clinical trials for invasive bladder cancer. Novel treatment approaches are required to improve survival further in patients with apparently localized disease.

Acute Toxicity in High-Risk Prostate Cancer Patients Treated With Androgen Suppression and Hypofractionated Intensity-Modulated Radiotherapy

PURPOSE To report acute toxicity resulting from radiotherapy (RT) dose escalation and hypofractionation using intensity-modulated RT (IMRT) treatment combined with androgen suppression in high-risk prostate cancer patients. METHODS AND MATERIALS Sixty patients with a histological diagnosis of high-risk prostatic adenocarcinoma (having either a clinical Stage of > or =T3a or an initial prostate-specific antigen [PSA] level of > or =20 ng/ml or a Gleason score of 8 to 10 or a combination of a PSA concentration of >15 ng/ml and a Gleason score of 7) were enrolled. RT prescription was 68 Gy in 25 fractions (2.72 Gy/fraction) over 5 weeks to the prostate and proximal seminal vesicles. The pelvic lymph nodes and distal seminal vesicles concurrently received 45 Gy in 25 fractions. The patients were treated with helical TomoTherapy-based IMRT and underwent daily megavoltage CT image-guided verification prior to each treatment. Acute toxicity scores were recorded weekly during RT and at 3 months post-RT, using Radiation Therapy Oncology Group acute toxicity scales. RESULTS All patients completed RT and follow up for 3 months. The maximum acute toxicity scores were as follows: 21 (35%) patients had Grade 2 gastrointestinal (GI) toxicity; 4 (6.67%) patients had Grade 3 genitourinary (GU) toxicity; and 30 (33.33%) patients had Grade 2 GU toxicity. These toxicity scores were reduced after RT; there were only 8 (13.6%) patients with Grade 1 GI toxicity, 11 (18.97%) with Grade 1 GU toxicity, and 5 (8.62%) with Grade 2 GU toxicity at 3 months follow up. Only the V60 to the rectum correlated with the GI toxicity. CONCLUSION Dose escalation using a hypofractionated schedule to the prostate with concurrent pelvic lymph node RT and long-term androgen suppression therapy is well tolerated acutely. Longer follow up for outcome and late toxicity is required.

Tumor invasion margin on the Riemannian space of brain fibers

Glioma is one of the most challenging types of brain tumors to treat or control locally. One of the main problems is to determine which areas of the apparently normal brain contain glioma cells, as gliomas are known to infiltrate several centimeters beyond the clinically apparent lesion that is visualized on standard Computed Tomography scans (CT) or Magnetic Resonance Images (MRIs). To ensure that radiation treatment encompasses the whole tumor, including the cancerous cells not revealed by MRI, doctors treat the volume of brain that extends 2cm out from the margin of the visible tumor. This approach does not consider varying tumor-growth dynamics in different brain tissues, thus it may result in killing some healthy cells while leaving cancerous cells alive in the other areas. These cells may cause recurrence of the tumor later in time, which limits the effectiveness of the therapy. Knowing that glioma cells preferentially spread along nerve fibers, we propose the use of a geodesic distance on the Riemannian manifold of brain diffusion tensors to replace the Euclidean distance used in the clinical practice and to correctly identify the tumor invasion margin. This mathematical model results in a first-order Partial Differential Equation (PDE) that can be numerically solved in a stable and consistent way. To compute the geodesic distance, we use actual Diffusion Weighted Imaging (DWI) data from 11 patients with glioma and compare our predicted infiltration distance map with actual grwoth in follow-up MRI scans. Results show improvement in predicting the invasion margin when using the geodesic distance as opposed to the 2cm conventional Euclidean distance.

Time course of prostatic edema post permanent seed implant determined by magnetic resonance imaging

PURPOSE To quantify the time course of postimplant prostatic edema magnitude and spatial isotropy using serial magnetic resonance imaging (MRI). METHODS AND MATERIALS Forty patients with histologic diagnosis of prostate cancer received an iodine-125 seed implant (Day 0) and consented to 1.5-T MRI on Days -1, 0, 14, and 28. Seeds of strength 0.39mCi were placed in a modified peripheral loading pattern to deliver 145Gy to the target volume. MR images consisted of 3-4mm thick axial slices with no gap. The image sets were anonymized and randomized to minimize contouring bias, then contoured by a single radiation oncologist. Contours were reoriented about their center of mass to align the prostate long axis with the superior-inferior (S-I) direction; prostate volumes and dimensions in the left-right (L-R), anterior-posterior (A-P), and S-I directions through the center of mass were calculated. RESULTS The average relative edema volume was 1.18±0.14 (1standard deviation) on Day 0 and 1.01±0.15 on Day 30. Between Days 0 and 30, the edema resolved linearly with time on average. Average relative edema dimensions on Day 0 in the L-R, A-P, and S-I directions were 1.01±0.07, 1.11±0.09, and 1.08±0.13, respectively. CONCLUSIONS As measured using MRI, the average edema magnitude for our study population was ∼20% on Day 0 and resolved linearly with time to ∼0% on Day 30. The edema exhibited spatial anisotropy, the prostate expanding on Day 0 by ∼10% in each of the A-P and S-I directions and by ∼0% in the L-R direction.