To estimate the Alpha/Beta (α/β) ratio for radiation(RT)-induced lung injury by comparing the density changes observed in patients treated with hypofractionated stereotactic body radiation therapy (SBRT) versus those treated with conventionally-fractionated (i.e., 2 Gy/fraction) radiation. Images from patients treated with hypofractionated SBRT (48 Gy in 12 fractions through multiple non-axial beams) were compared with the corresponding images from patients treated with conventionally-fractionated radiation treatment (usually with large fields- anteroposterior/posteroanterior and off-cord-obliques, either sequentially or concurrently). For all patients, the pre- and post-computed tomography (CT) images were compared to each other and to the three-dimensional dose distribution, using image fusion and treatment planning software. Therefore, we were able to define changes in regional lung density within different regions of the lung that received different doses of radiation (and hence different radiation fraction sizes). Data from both treatment cohorts was pooled for analysis. Linear regression was used to estimate the α/β ratio by comparing the observed dose response curve (DRC) data in the hypofractionated vs. conventionally-fractionated patients. Considering the data <6 months post-RT, the degree of density change was greater in the hypofractionated patients vs. conventionally-fractionated patients across the entire dose spectrum. Those treated with conventional fractionation, the slope of the dose response curve for radiation-associated density changes was ≈ 1.0, vs. a slope of ≈2.2 those treated with hypofractionated SBRT. Different values for α/β were estimated for different regions of the lung exposed to different total doses of radiation. For lungs exposed to total doses 12, 24, 36, and 48 Gy, the estimated values for α/β were 1.3, 2.5, 3.7, and 5.0, respectively. The estimated α/β ratio for acute RT-induced changes in lung density (assessed by CT) is in the range of 1.3-5.0, and appears to vary with total dose. The lower α/β ratio in the low dose region might reflect the lesser degree of injury observed with these lung doses. Additional work is needed to better define the fractionation sensitivity-associated lung injury, and also the impact of total dose on the fractionation sensitivity. This information might be useful in computing iso-effective dose distributions for competing treatment plans with variable fraction sizes, and for predicting subsequent degree of RT-associated lung injury.

Initial management of advanced NSCLC is typically systemic chemotherapy; thoracic radiation (RT) is usually reserved for palliation of focal symptoms. The rate of progressive intrathoracic disease leading to lobar collapse and/or urgent thoracic RT (events), and clinical factors associated with such progression, are not well described. We report a secondary analysis of a prospective clinical trial using first-line chemotherapy in advanced NSCLC. From 1997-2001, 67 patients with advanced IIIB (malignant pleural effusion or advanced supraclavicular adenopathy) or stage IV NSCLC were enrolled onto a phase III randomized trial comparing four cycles vs. continuous carboplatin/paclitaxel chemotherapy. The overall survival, response rates, and quality of life were not significantly different in the two arms. We retrospectively reviewed clinical and radiographic data, including lung imaging studies (CXR and CT) from enrollment until death to assess for new lobar collapse, and clinical records to identify receipt of urgent RT. For patients with atelectasis of a lobe at presentation (16 patients), collapse of a different lobe was required to score as an event. Univariate analyses using Fisher's exact test were used to assess the impact of various clinical factors on the rate of new lobar collapse and/or urgent RT. Sixty-seven pts were analyzed; 84% were Stage IV and 16% advanced Stage IIIB. Fifty-eight had subsequent chest imaging available for review. The median survival for all patients was 6.2 months. Twenty events were recorded (8 patients developed new lobar collapse, 6 patients received RT, and 6 patients had both). The fourteen new collapses were single lobe in 7 and whole lung in 7. Urgent RT was delivered for symptoms of dyspnea (9), dysphagia (2), hemoptysis (2), and pain (3). The median time from enrollment to event was 3.9 months. Amongst all 67 patients, those with hilar nodes at diagnosis had a 39% incidence of a subsequent thoracic event vs. 19% for patients without hilar nodes (p = .055). Approximately one third of patients receiving first line chemotherapy for NSCLC subsequently experienced a radiologically confirmed new lobar collapse or required urgent thoracic RT. Patients with hilar nodes were particularly at risk of a subsequent thoracic event. These results suggest that “early” thoracic RT may be helpful in preventing/delaying progressive intrathoracic disease in selected patients. Additional study is needed to better define the utility of “early” RT in this setting.

Radiation oncologists have a tradition of peer-review for quality assurance (QA). For example, “chart rounds” (the systematic review of patient's charts and films early during the course of therapy) are common in our field. There may be a need to refine our QA procedures: evolving RT techniques are increasingly complex; and medical records contain far more/complex information. The concept of “chart rounds” can be questioned as medical records become electronic and beams/doses/volumes are increasingly difficult to systematically review. We herein quantitatively assess the utility of physician peer review at different points of the patient's continuum of care. Our department has a two-step physician peer-review process. First, daily meetings to prospectively review the planned doses and target/image segmentation prior to RT planning/delivery. Second, weekly “traditional chart rounds” where port/setup films, 3D RT doses, and chart documentation are reviewed. These QA activities are attended by 3–6 faculty physicians, 2–4 residents, and representatives from dosimetry, physics, and therapy. During a 6-week period, we monitored the utility of the peer-review process. Each reviewed case was scored as No Change, Minor Change (e.g., 0.5 cm modification to segmented volume), or Major Change (e.g., extension of target volume to include another nodal region or site, >1 cm change to a gross target volume). The results from the two different peer review activities were summarized by descriptive statistics; 2-tailed Fisher's exact test for comparisons. A total of 105 cases had daily pre-RT review. Of these, 17% were No Changes, 28% were Minor Changes, and 55% were Major Changes. In the “weekly traditional chart rounds”, 108 cases were reviewed. Of these, 92% were No Changes, 5% a Minor Changes, and 3% a Major Changes. The rates of Major and Minor Changes were higher for the pre-RT daily review than the traditional weekly chart rounds (p < 0.001). Daily peer-review prior to RT identifies many more minor and major changes than daily weekly “chart rounds”. This is expected as issues identified in the daily pre-RT peer review session usually are addressed prior to the latter peer review. Indeed, the low rate of changes in the second peer-review (i.e., weekly “chart rounds”) demonstrates the effectiveness of the initial pre-RT peer review. Since the initial review is done PRIOR to dosimetry/physics-based treatment planning, this saves dosimetry/physics effort as modification to target volumes are largely made prior to (often time-consuming IMRT) planning. Additional analysis is underway to better understand the types and sources of peer-review-defined changes and the cost-effectiveness of such peer-review activities.

7528 Background: Therapies directed at both the epidermal growth factor receptor (EGFR) and the vascular endothelial growth factor (VEGF) pathways have been shown to improve survival in NSCLC and also have radiosensitizing properties. Methods: Pts receive Ind Cb (AUC 6), P (225 mg/m 2 ) and B (15 mg/kg) on d1 and 22. PET scans are done pre- and post-I. On day 43, pts receive weekly Cb (AUC 2 x 7) and P (45 mg/m 2 x 7) with 74 Gy (2 Gy/d) of thoracic conformal radiotherapy (TCRT). Cohort I (n=5) received B at 10 mg/kg q2wks during C therapy. Cohorts II and III (both n=5) received the same dose of B as in cohort I but also received Er at 100 mg and 150 mg po Tuesday - Friday of each week of C therapy, respectively. The primary endpoint is PFS at 1 year. All histologies are allowed including squamous (SQ) (an early stopping rule is in place for pulmonary hemorrhagic (PH) complications in SQ pts). Results: Thus far, 31 eligible PS 0–1 pts have been accrued (med age 62 yrs, range 41–74, 19 non- squamous, 12 SQ, 63% IIIA, 37% IIIB). Ind CbP + B has been well tolerated (1 gr 3 hypertension). No PH during Ind has been seen (including the 12 SQ cell pts). Response after Ind, 37% PR, 59% SD, 4% PD. Tumor volumes and PET SUVs have significantly decreased comparing pre- and post-Ind studies (p=0.0001 and p=0.0002, respectively). Cohort II has been expanded as the phase II regimen. To date, 25 of 26 (96%) pts have achieved the dose of 74 Gy (1 pt stopped at 60 Gy due to ILD). During Conc therapy, the principal toxicity has been esophagitis (53.8% gr 2, 19.2% gr 3). One grade 3 PH occurred in 1 SQ pt. One gr 5 late (> 2 mos after treatment) PH occurred in a SQ pt. Overall response rate following treatment - 68.2% (95% CI, 45–86%). The PFS at 1 year is 58% (95% CI, 34–76%) with an estmated 1-year overall survival rate of 79% (95% CI, 53–92%) which compares favorably to our historical experience. Conclusions: Preliminarily, we conclude that 1) Incorporation of B and E into this treatment paradigm appears feasible, 2) Esophagitis remains the primary toxicity, 3) Phase II accrual continues but early analysis of survival appears promising. Further details regarding the TCRT parameters and toxicity will be presented. [Table: see text]

Difficult economic circumstances have renewed interest in the cost effectiveness of medical interventions. In the setting of cancer metastatic to the brain, a condition in which few patients experience long term survival, at least three different radiation treatment paradigms are employed. Either SRS alone or SRS in combination with WBRT is commonly selected for 1–4 lesions whereas WBRT alone is frequently reserved for more numerous metastases. The purpose of this study is to present both an analysis of cost-effectiveness for these three paradigms and a cost-utility model for SRS alone versus SRS plus WBRT. A retrospective analysis of patients recently treated with Cyberknife SRS was undertaken to gain real figures of the charges and reimbursements for both technical and professional activities surrounding its use. A similar review obtained data for WBRT. Only direct costs to the health care system were included in this analysis, such as costs of the primary treatment, costs of salvage, and costs of follow-up imaging. Overall survival data was obtained from randomized trials if available, specifically RTOG 9508 (Andrews trial) for WBRT alone and WBRT and SRS, the Aoyama trial for SRS alone, and historical data for observation and steroid management only. A model of quality adjusted survival after SRS alone versus SRS plus WBRT was developed from the Aoyama trial mini mental status exam data. Finally, a sensitivity analysis is included to test the reliability of the data. Mean total costs for SRS alone, WBRT alone, and SRS plus WBRT, including follow-up imaging costs and the cost of salvage treatment, amounted to $18,610, $9,610, and $23,476, respectively. For patients with one brain metastasis, the cost per life year compared to observation alone totaled $56K for SRS plus WBRT, $45K for SRS alone, and $34K for WBRT alone. For 2–4 metastases, SRS alone, WBRT plus SRS and WBRT alone cost $66K, $83K, and $34K per life year, respectively. WBRT for greater than 4 metastases costs $34K per life year. The overall cost effectiveness of SRS alone versus SRS plus WBRT differ relatively little for less than 4 metastases because the predominant determinant of the cost/benefit ratio is the RTOG RPA class of the patient rather than the choice of modality. The use of SRS for one lesion is more cost-effective than for 2–4 metastases, whether or not WBRT is additionally employed.

20738 Background: Elderly patients with glioblastoma (GBM) have poor outcomes. Short course radiation improves median survival over best supportive care (10 vs 4 months). Standard care for patients <70 years old includes radiation and concurrent temozolomide followed by adjuvant temozolomide (i.e., the “Stupp regimen”). Since 2002 we have offered this regimen as an alternative to select elderly patients with GBM. However, there is no data limited to the over 70 population using this approach. We undertook this IRB approved chart review to describe our outcomes with this approach. Methods: We reviewed the records of all patients over age 70 diagnosed with GBM since 2002 at the University of North Carolina to determine treatment received, performance status, age at diagnosis, and RPA class. Median survival was calculated according to the Kaplan-Meier method. Results: We identified 26 patients with a median survival of 20.6 weeks. One patient was excluded due to incomplete follow-up. Thirteen patients received best supportive care with a median survival of 8.6 weeks. Eight patients were treated with the Stupp regimen, 3 patients with short course radiation, and 2 with high dose radiation without chemotherapy; median survival was 38.0, 32.3, and 23.1 weeks, respectively. Median survival for patients classified as RPA class V and VI (36.9 vs 10.3 weeks) was consistent with that previously reported by the RTOG. Patients aged 70–80 fared better than those > 80 (22.1 vs 11.0 weeks, respectively). Increased performance status correlated with improved survival (Pearson r = 0.57, p=0.0024). Conclusions: Offering aggressive treatment with the Stupp regimen to good performance status elderly patients with GBM is a reasonable approach. In our experience, concurrent therapy with temozolomide and radiation is well tolerated with an overall survival slightly better than short course radiation alone. In our opinion, a frank discussion with good performance status elderly patients and their families regarding treatment options can include chemoradiation. Median survival by treatment, performance status, and age (weeks, n) All patients PS <70 PS ≥70 Age 70–80 Age >80 Hospice/supportive care 8.6 (13) 8.4 (11) 9.6 (2) 8.4 (8) 10.3 (5) Short course RT (4000 cGy) 32.3 (3) 34.5 (2) 32.3 (1) 34.5 (2) 32.3 (1) Long course RT (6000 cGy) 23.1 (2) (0) 23.1 (2) 23.1 (2) (0) Concurrent radiation (6120 cGy) and temozolomide 38.0 (8) 42.4 (2) 38 (6) 36.9 (7) 38 (1) No significant financial relationships to disclose.

7145 Background: Combined modality treatment is the standard of care for patients (pts.) with unresectable stage III NSCLC. Dose escalation of radiotherapy and the use of concurrent chemotherapy are two strategies attempting to improve survival and locoregional control. The intensification of therapy increases the risk of both early and late treatment related toxicities. Methods: From 5/1996 to 8/2004, 112 stage III NSCLC pts. were entered into 4 Phase I/II trials to assess the safety and feasibility of high-dose (74–90 Gy) thoracic conformal radiotherapy (TCRT) in QD or BID fractions. All pts. were treated with platinum-based induction chemotherapy; 3 of the trials also used concurrent chemotherapy. Results: The median follow up of survivors (29/112) on these trials was 4.9 years. The overall response rate after combined modality therapy was 47% (53/112) (CR 4%, 5/112; PR 43%, 48/112). 27% (30/112) had stable disease. The median survival (with 95% CI) was 24 months (18–31 months). 1-, 3-, and 5-year overall survival was 69% (60–77%), 36% (27–45%), and 24% (16–33%) respectively. Late complications of therapy (defined as >90 days post radiotherapy reported to date) are displayed in the table. Two pts. developed a second primary (1 lung, 1 liver carcinoid). In total, 22% (25/112) had late complications. These patients appear to have a significantly better overall survival (p = .007). 12% (13/112) had a brain-only recurrence, although this did not seem to significantly impact overall survival (p = .82). Conclusions: 1) High-dose TCRT is feasible and results in promising survival outcomes. 2) Late complications occur in a minority of patients suggesting the potential benefit of more aggressive TCRT is not outweighed by its risk. 3) Interestingly, brain-only recurrences did not significantly impact survival in these trials. [Table: see text] [Table: see text]

Purpose/Objective: Surgeons generally do not review preoperative radiotherapy (RT) fields when planning for esophagogastrectomy. Because no studies have detailed anastomotic leak or stricture rates relative to whether the esophageal or gastric transection sites were within or outside the radiation field, we reviewed this in our experience (January 1996 to July 2004). Materials/Methods: The extent of radiation fields relative to the carina was determined from port films. An independent reviewer determined the position of the anastomosis from postoperative films. Neither reviewer was made aware of the occurrence of an anastomotic leak or stricture. Results: The overall leak rate in 67 esophagogastrectomy patients was 15%(10/67). 75%(50/67) received preoperative RT (40–45 Gy), of which 38 had radiation port films available for review. Leaks occurred in 14%(7/50) of RT patients and 18%(3/17) of non-RT patients. The leak rate was the same whether or not the radiation field encompassed the esophageal transection site (14%3/22 vs. 13%2/16), but tended to be higher if the fundus was radiated (15%4/26 vs. 8%1/12 p=NS) or if both the esophageal and fundic transection sites were radiated (23%3/13 vs. 8%2/25 p=NS). Anastomotic stricture occurred in 16%(11/67) of patients, of which 5 had received RT (encompassing both gastric and esophageal transaction sites in all). Conclusions: The leak rate is the same for RT vs. non-RT patients, and is similar whether or not the esophageal side is radiated, but appears to be higher if the gastric or both the gastric and esophageal margins have been irradiated.

Management of advanced-stage uterine serous carcinoma (USC) is uncertain, and postsurgical therapeutic options swing between radiation and chemotherapy. The aim of this study is to evaluate the utility of radiotherapy compared to platinum-based chemotherapy in women with advanced-stage USC. We retrospectively identified cases of USC at our institution. Survival distributions were calculated by the Kaplan–Meier method. Two-tailed t-tests were used to compare time to progression and time to death. We identified 24 women diagnosed with either stage III or IV USC. Time to progression for women receiving radiotherapy was 5.3 months as compared with 12.4 months for women receiving chemotherapy (P = 0.01). Mean time to death for the radiotherapy group was 8 months compared to 18 months in the chemotherapy group (P = 0.04). Kaplan–Meier survival curves were significantly different between the two groups (P = 0.01). While radiotherapy appears to control USC recurrences in the pelvis, the disease often recurs distantly. When compared to radiotherapy, platinum-based chemotherapy appears to increase disease-free survival and time to death in women with advanced-stage USC.

Purpose/Objective: We performed a systematic review and meta-analysis of chemotherapy alone compared to chemotherapy plus radiation in the management of adult early stage aggressive non-Hodgkin's lymphoma (NHL). Overall survival (OS) and disease-free survival (DFS) at 5 years were evaluated.

Purpose/Objective: The appropriate combined modality approach to treat Stage IIIA/B non-small cell lung cancer remains unclear, but control of both local and occult metastatic disease remains problematic. The objectives are: 1. To escalate the dose of 3-D conformal radiation therapy with concurrent paclitaxel and carboplatin from 78 Gy to 90 Gy and describe the toxicity and maximum tolerated dose of conformal radiation; 2. To assess the toxicity and efficacy of two cycles of induction paclitaxel/carboplatin/irinotecan with filgrastim in stage IIIA/B non-small cell lung cancer; 3. To assess recurrence patterns in this patient population; and, 4. To assess the response rate and survival of patients with stage IIIA/B non-small cell lung cancer managed in this fashion. Materials/Methods: Intensified induction therapy (carboplatin (C) AUC=5, irinotecan (I) 100mg/m2, paclitaxel (P) 175 mg/m2 on d1 and 22 with filgrastim) followed by dose-escalated 3-D radiation therapy beginning on d43 at 78Gy, 82Gy, 86Gy, and 90 Gy with weekly carboplatin (C) AUC=2 and paclitaxel (P) 45 mg/m2. The elective mediastinum was treated to doses of 40–50 Gy. Radiation was delivered in 2Gy daily fractions. Results: Demographics:Twenty-nine ECOG 0–1 patients have been enrolled (19M:10F; median age 59 (40-81); 13 IIIA, 16 IIIB; 62% ≥5% wt. loss; median FEV11.7 (0.8–4.2)). Induction chemotherapy toxicity:Hematologic toxicity of chemotherapy has been minimal with 1 cycle complicated by gr 3 fever/neutropenia. Non-hematologic toxicity was also minimal with 1 case of gr 3 fatigue, nausea, and emesis. Response to induction CIP chemo: 46% PR, 54% SD, 0% PD (25 of 27 are evaluable thusfar); Median pre-chemo gross tumor volume (GTV)=66cc (13–247); Median post-chemo GTV=28cc (7–247). Radiation treatment details:Thus far, 25 of 27 pts (93%) have started radiation on day 43. The V20was less than 35% in all pts except 2 where it was 40% and 38%, respectively. Median length of esophagus receiving 40 Gy (L40) and 60 Gy (L60) was 12cm(8.3–19cm) and 6.5cm(0–16.5cm). Three patients who started radiation failed to complete the prescribed dose. 78 Gy dose level: Five pts were treated at the 78 Gy level with 1 gr 3 esophagitis and 1 gr 3 broncho-esophageal fistula. The patient with the fistula had treatment discontinued at 74 Gy and subsequently died related to a stent complication. 82 Gy dose level:Seven pts were treated at 82 Gy with 1 gr 3 esophagitis. One late toxicity of bronchial stenosis gr 2 was observed. One patient had fatal pulmonary hemoptysis with no evidence of disease on autopsy one year after completion of treatment. One patient died in week 1 of radiation due to an embolic event unrelated to treatment. 86 Gy dose level:Seven pts were treated at 86 Gy with 1 gr 3 esophagitis and 1 gr 3 fatigue requiring discontinuation of therapy at 70 Gy. One late toxicity of gr 2 bronchial stenosis was seen. 90 Gy dose level:Six patients have entered the 90 Gy dose level with 4 completing therapy (2 remain on tx.) with no ≥gr 3 toxicity identified. Survival and Response to therapy:The one year survival probability is 0.79. One local progression has been observed in a pt who received 82 Gy. Five patients have developed distant failures with 3 developing brain metastases. All evaluable patients at the 90 Gy level have obtained a complete response by either bronchoscopy or radiographic imaging. Conclusions: (1.) Induction with the drug triplet of CIP is active and well tolerated. (2.) Utilizing 3-D techniques, doses of thoracic radiation in excess of 86Gy can be administered with concurrent chemotherapy after aggressive induction therapy with acceptable acute toxicity. (3.) The impact of this aggressive approach on recurrence patterns, late toxicity, and survival will require further follow-up.

Three-Dimensional conformal radiation treatment (3D-CRT) planning and delivery is an external beam radiation therapy modality that has the general goal of conforming the shape of a prescribed dose volume to the shape of a 3-dimensional target volume, simultaneously limiting dose to critical normal structures. 3-Dimensional conformal therapy should include at least one volumetric imaging study of the patient. This image should be obtained in the treatment position for visualizing the target and normal anatomic structures that are potentially within the irradiated volume. Most often, computed tomography (CT) and/or magnetic resonance imaging (MRI) are used; however, recently, other imaging modalities such as functional MRI, MR spectroscopy, and positron emission tomography (PET) scans have been used to visualize the clinically relevant volumes. This article will address the clinically relevant issues with regard to low- and intermediate-grade gliomas and the role of 3D-CRT planning. Specific issues that will be addressed will include normal tissue tolerance, target definition, treatment field design in regard to isodose curves and dose-volume histograms, and immobilization.