John M. Buatti

3D Slicer as an image computing platform for the Quantitative Imaging Network

Quantitative analysis has tremendous but mostly unrealized potential in healthcare to support objective and accurate interpretation of the clinical imaging. In 2008, the National Cancer Institute began building the Quantitative Imaging Network (QIN) initiative with the goal of advancing quantitative imaging in the context of personalized therapy and evaluation of treatment response. Computerized analysis is an important component contributing to reproducibility and efficiency of the quantitative imaging techniques. The success of quantitative imaging is contingent on robust analysis methods and software tools to bring these methods from bench to bedside. 3D Slicer is a free open-source software application for medical image computing. As a clinical research tool, 3D Slicer is similar to a radiology workstation that supports versatile visualizations but also provides advanced functionality such as automated segmentation and registration for a variety of application domains. Unlike a typical radiology workstation, 3D Slicer is free and is not tied to specific hardware. As a programming platform, 3D Slicer facilitates translation and evaluation of the new quantitative methods by allowing the biomedical researcher to focus on the implementation of the algorithm and providing abstractions for the common tasks of data communication, visualization and user interface development. Compared to other tools that provide aspects of this functionality, 3D Slicer is fully open source and can be readily extended and redistributed. In addition, 3D Slicer is designed to facilitate the development of new functionality in the form of 3D Slicer extensions. In this paper, we present an overview of 3D Slicer as a platform for prototyping, development and evaluation of image analysis tools for clinical research applications. To illustrate the utility of the platform in the scope of QIN, we discuss several use cases of 3D Slicer by the existing QIN teams, and we elaborate on the future directions that can further facilitate development and validation of imaging biomarkers using 3D Slicer.

A multiinstitutional outcome and prognostic factor analysis of radiosurgery for resectable single brain metastasis

PURPOSE Recent randomized trials of selected patients with single brain metastasis comparing resection followed by whole-brain radiotherapy (WBRT) to WBRT alone have shown a statistically significant survival advantage for surgery and WBRT. A multiinstitutional retrospective study was performed, which identified comparable patients who were treated with stereotactic radiosurgery (RS) and WBRT. METHODS AND MATERIALS The RS databases of four institutions were reviewed to identify patients who met the following criteria: single-brain metastasis; no prior cranial surgery or WBRT; age > 18 years; surgically resectable lesion; Karnofsky Performance Status (KPS) > or = 70 at time of RS; nonradiosensitive histology. One hundred twenty-two patients were identified who met these criteria. Patients were categorized by: (a) status of the primary, (b) status of non-CNS metastasis, (c) age, (d) baseline KPS (from 70-100), (e) histology, (f) time from diagnosis of primary to the detection of the brain metastasis, (g) gender, and (h) tumor volume. RS was performed with a linear accelerator based technique (peripheral dose range was 10-27 Gy, median was 17 Gy). WBRT was performed in all but five patients who refused WBRT (dose range was 25-40 Gy, median was 37.5 Gy). RESULTS The median follow-up for all patients was 123 weeks. The overall local control rate (defined as lack of progression in the RS volume) was 86%. Intracranial recurrence outside of the RS volume was seen in 27 patients (22%). The actuarial median survival from date of RS is 56 weeks, and the 1-year and 2-year actuarial survival rates are 53% and 30%. The median duration of functional independence (sustained KPS > or = 70) is 44 weeks. Nineteen of 77 deaths were attributed to CNS progression (25% of all deaths). Multivariate analysis revealed the following factors to be statistically significant predictors of survival: baseline KPS (p < .0001) and absence of other sites of metastasis (p = 0.008). CONCLUSION The RS in conjunction with WBRT for single brain metastasis can produce substantial functional survival, especially in patients with good performance status and without extracranial metastasis. These results are comparable to recent randomized trials of resection and WBRT. The advantages of RS over surgery in terms of cost, hospitalization, morbidity, and wider applicability strongly suggest that a randomized trial to compare RS with surgery is warranted.

A multi-institutional review of radiosurgery alone vs. radiosurgery with whole brain radiotherapy as the initial management of brain metastases

PURPOSE Data collected from 10 institutions were reviewed to compare survival probabilities of patients with newly diagnosed brain metastases managed initially with radiosurgery (RS) alone vs. RS + whole brain radiotherapy (WBRT). METHODS AND MATERIALS A database was created from raw data submitted from 10 institutions on patients treated with RS for brain metastases. The major exclusion criteria were resection of a brain metastasis and interval from the end of WBRT until RS >1 month (to try to ensure that the up-front intent was to combine RS + WBRT and that RS was not given for recurrent brain metastases). Survival was estimated using the Kaplan-Meier method from the date of first treatment for brain metastases until death or last follow-up. Survival times were compared for patients managed initially with RS alone vs. RS + WBRT using the Cox proportional hazards model to adjust for known prognostic factors or Radiation Therapy Oncology Group recursive partitioning analysis (RPA) class. RESULTS Out of 983 patients, 31 were excluded because treatment began after 6/1/98; 159 were excluded because brain metastases were resected; 179 were excluded because there was an interval >1 month from WBRT until RS; and 45 were excluded for other reasons. Of the 569 evaluable patients, 268 had RS alone initially (24% of whom ultimately had salvage WBRT), and 301 had RS + up-front WBRT. The median survival times for patients treated with RS alone initially vs. RS + WBRT were 14.0 vs. 15.2 months for RPA Class 1 patients, 8.2 vs. 7.0 months for Class 2, and 5.3 vs. 5.5 months for Class 3, respectively. With adjustment by RPA class, there was no survival difference comparing RS alone initially to RS + up-front WBRT (p = 0.33, hazard ratio = 1.09). CONCLUSIONS Omission of up-front WBRT does not seem to compromise length of survival in patients treated with RS for newly diagnosed brain metastases.

Benign meningiomas: primary treatment selection affects survival.

PURPOSE To examine the effect of primary treatment selection on outcomes for benign intracranial meningiomas at the University of Florida. METHODS AND MATERIALS For 262 patients, the impact of age, Karnofsky performance status, pathologic features, tumor size, tumor location, and treatment modality on local control and cause-specific survival was analyzed (minimum potential follow-up, 2 years; median follow-up, 8.2 years). Extent of surgery was classified by Simpson grade. Treatment groups: surgery alone (n = 229), surgery and postoperative radiotherapy (RT) (n = 21), RT alone (n = 7), radiosurgery alone (n = 5). Survival analysis: Kaplan-Meier method with univariate and multivariate analysis. RESULTS At 15 years, local control was 76% after total excision (TE) and 87% after subtotal excision plus RT (SE+RT), both significantly better (p = 0.0001) than after SE alone (30%). Cause-specific survival at 15 years was reduced after treatment with SE alone (51%), compared with TE (88%) or SE+RT (86%) (p = 0.0003). Recurrence after primary treatment portended decreased survival, independent of initial treatment group or salvage treatment selection (p = 0.001). Atypical pathologic features predicted reduced 15-year local control (54 vs. 71%) and cause-specific survival rates (57 vs. 86%). Multivariate analysis for cause-specific survival revealed treatment group (SE vs. others; p = 0.0001), pathologic features (atypical vs. typical;p = 0.0056), and Karnofsky performance status (> or = 80 vs. < 80; p = 0.0153) as significant variables. CONCLUSION Benign meningiomas are well managed by TE or SE+RT. SE alone is inadequate therapy and adversely affects cause-specific survival. Atypical pathologic features predict a poorer outcome, suggesting possible benefit from more aggressive treatment. Because local recurrence portends lower survival rates, primary treatment choice is important.

RADIOSURGERY FOR PATIENTS WITH BRAIN METASTASES: A MULTI-INSTITUTIONAL ANALYSIS, STRATIFIED BY THE RTOG RECURSIVE PARTITIONING ANALYSIS METHOD

PURPOSE To estimate the potential improvement in survival for patients with brain metastases, stratified by the Radiation Therapy Oncology Group (RTOG) recursive partitioning analysis (RPA) class and treated with radiosurgery (RS) plus whole brain radiotherapy (WBRT). METHODS AND MATERIALS An analysis of the RS databases of 10 institutions identified patients with brain metastates treated with RS and WBRT. Patients were stratified into 1 of 3 RPA classes. Survival was evaluated using Kaplan-Meier estimates and proportional hazard regression analysis. A comparison of survival by class was carried out with the RTOG results in similar patients receiving WBRT alone. RESULTS Five hundred two patients were eligible (261 men and 241 women, median age 59 years, range 26-83). The overall median survival was 10.7 months. A higher Karnofsky performance status (p = 0.0001), a controlled primary (median survival = 11.6 vs. 8.8 months, p = 0.0023), absence of extracranial metastases (median survival 13.4 vs. 9.1 months, p = 0.0001), and lower RPA class (median survival 16.1 months for class I vs. 10.3 months for class II vs. 8.7 months for class III, p = 0.000007) predicted for improved survival. Gender, age, primary site, radiosurgery technique, and institution were not prognostic. The addition of RS boosted results in median survival (16.1, 10.3, and 8.7 months for classes I, II, and III, respectively) compared with the median survival (7.1, 4.2, and 2.3 months, p <0.05) observed in the RTOG RPA analysis for patients treated with WBRT alone. CONCLUSION In the absence of randomized data, these results suggest that RS may improve survival in patients with BM. The improvement in survival does not appear to be restricted by class for well-selected patients.

Radiotherapy for pituitary adenoma: Long-term outcome and sequelae☆

PURPOSE To review outcome and treatment sequelae in patients treated with external beam radiotherapy for pituitary adenomas. METHODS AND MATERIALS One hundred forty-one patients with pituitary adenomas received radiotherapy at the University of Florida and had 2-year minimum potential follow-up. One hundred twenty-one had newly diagnosed adenomas, and 20 had recurrent tumors. Newly diagnosed tumors were treated with surgery and radiotherapy (n = 98) or radiotherapy alone (n = 23). Patients with recurrent tumors received salvage treatment with surgery and radiotherapy (n = 10) or radiotherapy alone (n = 10). The impact of age, sex, presenting symptoms, tumor extent, surgery type, degree of resection, hormonal activity, primary or salvage therapy, and radiotherapy dose on tumor control was analyzed. Tumor control is defined by the absence of radiographic progression and stable or decreased hormone level (in hormonally active tumors) after treatment. Effect of therapy on vision, hormonal function, neurocognitive function, life satisfaction, and affective symptoms were examined. A Likert categorical scale survey was used for assessment of neurocognitive, life satisfaction, and affective symptom status. Survey results from the radiotherapy patients were compared with a control group treated with transsphenoidal surgery alone. Multivariate analysis used the forward step-wise sequence of chi squares for the log rank test. RESULTS At 10 years, tumor control for the surgery and radiotherapy group (S + RT) was 95% and not statistically different (p = 0.58) than for patients treated with radiotherapy alone (RT) (90%). Patients with prolactin- and ACTH-secreting tumors had significantly worse tumor control, as did patients treated for recurrent tumors. Multivariate analysis for tumor control revealed that only young age was predictive of worse outcome (p = 0.0354). Visual function was either unaffected or improved in most patients, although four patients developed visual loss due to treatment. Hormonal function was affected adversely in 46 of the 93 patients for whom detailed hormonal information was available. Neurocognitive function evaluation revealed that patients in the S + RT group were more likely (p = 0.005) to report difficulty with memory than those in the RT-alone or S-alone groups. No significant difference in life satisfaction or affective symptoms was evident. CONCLUSIONS Pituitary adenomas are well controlled by external beam radiotherapy, either alone or in combination with surgery. Visual symptoms often improve after treatment. Hormonal sequelae require medical intervention in many patients. Neurocognitive sequelae may be different among treatment groups.

Radiosurgery in the initial management of malignant gliomas: Survival comparison with the RTOG recursive partitioning analysis

PURPOSE To evaluate the impact of stereotactic radiosurgery on the survival of patients treated with malignant gliomas. METHODS AND MATERIALS A total of 115 patients from three institutions (75 from the Joint Center for Radiation Therapy, 30 from the University of Wisconsin, and 10 from the University of Florida) were treated with a combination of surgery, external beam radiation therapy, and linac-based radiosurgery as part of similar institutional protocols from March 1988 through July 1993. Patients were stratified into six prognostic classes (classes 1-6) based on the recursive partitioning analysis of multiple prognostic factors previously reported by the Radiation Therapy Oncology Group. RESULTS The actuarial 2-year and median survival for all patients analyzed was 45% and 96 weeks, respectively. In comparison to the results from a previously published analysis of 1578 patients entered on three Radiation Therapy Oncology Group external beam radiotherapy protocols from 1974 to 1989, those patients treated with radiosurgery had a significantly improved 2-year and median survival (p = 0.01) corresponding with a standardized mortality risk ratio of 0.51 [95% confidence interval (CI): 0.31, 0.85]. This improvement in survival was seen predominantly for the worse prognostic classes (classes 3-6). The 2-year survival for the radiosurgical patients compared with the previously reported patients was 81% vs. 76% for classes 1/2, 75% vs. 35% for class 3, 34% vs. 15% for class 4, and 21% vs. 6% for classes 5/6, respectively. Although Karnofsky performance status and prognostic class were significant on univariate analysis, only the Karnofsky score was a significant predictor of outcome on multivariate analysis. Median and 2-year survival for patients with a Karnofsky score > or = 70 was 106 weeks and 51%, respectively, as compared to 38 weeks and 0% for patients with a Karnofsky score < 70% (p = 0.001). CONCLUSIONS The addition of radiosurgery to conventional treatment (surgery and external beam radiotherapy) of malignant gliomas appears to improve survival when compared to historical reports. These results should be interpreted with caution because the recursive partitioning model does not completely predict the prognosis of the patients treated in the present study. Although this study suggests that radiosurgery may prolong survival in patients with malignant gliomas, the role of radiosurgery in the management of these patients remains to be defined by a prospective randomized trial.

Pixel analysis of MR perfusion imaging in predicting radiation therapy outcome in cervical cancer

The purpose of this study was to assess heterogeneity of tumor microcirculation determined by dynamic contrast‐enhanced magnetic resonance (MR) imaging and its prognostic value for tumor radiosensitivity and long‐term tumor control using pixel‐by‐pixel analysis of the dynamic contrast enhancement. Sixteen patients with advanced cervical cancer were examined with dynamic contrast‐enhanced MR imaging at the time of radiation therapy. Pixel‐by‐pixel statistical analysis of the ratio of post‐ to precontrast relative signal intensity (RSI) values in the tumor region was performed to generate pixel RSI distributions of dynamic enhancement patterns. Histogram parameters were correlated with subsequent tumor control based on long‐term cancer follow‐up (median follow‐up 4.6 years; range 3.8–5.2 years). The RSI distribution histograms showed a wide spectrum of heterogeneity in the dynamic enhancement pattern within the tumor. The quantity of low‐enhancement regions (10th percentile RSI < 2.5) significantly predicted subsequent tumor recurrence (88% vs. 0%, P = 0.0004). Discriminant analysis based on both 10th percentile RSI and pixel number (reflective of tumor size) further improved the prediction rate (100% correct prediction of subsequent tumor control vs. recurrence). These preliminary results suggest that quantification of the extent of poor vascularity regions within the tumor may be useful in predicting long‐term tumor control and treatment outcome in cervical cancer. J. Magn. Reson. Imaging 2000;12:1027–1033.

CURATIVE RADIOTHERAPY FOR PRIMARY ORBITAL LYMPHOMA

PURPOSE To review our institutional experience with primary orbital lymphoma and determine the prognostic factors for survival, local control, and distant metastases. In addition, we also analyzed the risk factors for complications in the radiotherapeutic management of this tumor. METHODS AND MATERIALS Between 1973 and 1998, 47 patients (29 women [62%] and 18 men [38%], median age 69 years, range 32-89) with Stage IAE orbital lymphoma were treated with curative intent at one department. Five had bilateral orbital involvement. The tumor was located in the eyelid and extraocular muscles in 23 (44%), conjunctiva in 17 (33%), and lacrimal apparatus in 12 (23%). The histologic features according to the World Heath Organization classification of lymphoid neoplasms was follicular lymphoma in 25, extranodal marginal zone B-cell lymphoma of mucosa-associated lymphoid tissue type in 8, diffuse large B-cell lymphoma in 12, mantle cell lymphoma in 6, and peripheral T-cell lymphoma in 1. For the purposes of comparison with the existing literature on orbital lymphomas, the grading system according to the Working Formulation was also recorded. The histologic grade was low in 33 (63%), intermediate in 18 (35%), and high in 1 (2%). All patients were treated with primary radiotherapy alone. The median dose for low-grade tumors was 3000 cGy (range 2000-4020); the median dose for intermediate and high-grade tumors was 4000 cGy (range 3000-5100). A lens-sparing approach was used in 19 patients (37%). Late complications for the lens and cornea were scored according to the subjective, objective, management, and analytic (SOMA) scale of the Late Effects of Normal Tissue (LENT) scoring system. The median follow-up was 55 months (range 6-232). RESULTS The local control rate was 100% in the 52 orbits treated. The 5-year overall survival and relapse-free survival rate was 73.6% and 65.5%, respectively. Tumor grade and location did not predict for overall survival or relapse-free survival. Seven patients (15%) developed distant recurrence (brain 2, extremity 2, mediastinum 1, liver 1, and retroperitoneum 1). One patient (2%) developed cervical node metastasis. The 5- and 10-year cataract-free survival rate was 56.7% and 32.9%, respectively. Of the 12 lens complications, 8 were LENT Grade 1 and 4 were Grade 3 toxicity. Only male gender predicted for an increased risk of cataract formation. Radiotherapy dose and technique did not predict for cataract formation; however, none of the patients who underwent the lens-sparing technique developed Grade 3 lens toxicity or required surgical correction. Of the nine corneal events, two were Grade 1, four Grade 2, and three were Grade 3 toxicity. Ten dry eyes were recorded; all were mild, and no patient had severe dry eye syndrome. Neovascular glaucoma was seen in 1 patient. No injury to the retina or optic nerve was reported. CONCLUSION Radiotherapy alone is a highly effective modality in the curative management of primary orbital lymphoma. Most complications were minimal and did not require medical or surgical intervention. Although the use of the lens-sparing technique did not influence the incidence of cataractogenesis, we continue to recommend this approach whenever possible, because our experience indicates a higher grade of toxicity occurs and a higher incidence of corrective surgery is needed in patients treated without lens protection.

Intracranial ependymomas: an analysis of prognostic factors and patterns of failure.

From 1965 to 1997, 49 patients were diagnosed and treated for intracranial ependymoma at one institution. Tumor location was infratentorial in two thirds, and pathology was low grade in 38 patients (78%). Gross total resection of the primary tumor was achieved in 21 (43%). Thirty-six patients received adjuvant radiotherapy; the entire neuraxis was treated in 14, whole brain in 10, and local field only in 12. Median follow-up was 9.6 years (range, 2–33 years). The 5-, 10-, and 15-year overall survival rates were 71.4%, 63.5%, and 63.5% for craniospinal radiotherapy, 60.0%, 60.0%, and 40.0% for whole brain radiotherapy, and 80.8%, 64.6%, and 64.6% for local field radiotherapy (p = 0.88). The 5-, 10-, and 15-year local control rates were 60.3%, 54.4%, and 48.9%. The prognostic factors for a better local control rate were gross total resection (p = 0.021) and low grade histology (p = 0.031). Seventeen of 43 (39.5%) M0 patients did not respond to treatment; all had local failure and 4 also had a spinal relapse. Spinal relapse developed in 3 of 31 (10%) M0 patients who did not receive spinal radiotherapy, whereas 1 of 12 (8%) who had spinal radiotherapy did not respond to treatment in the spine. The results of this study indicate that local radiotherapy is sufficient for M0 patients with intracranial ependymoma.