Search

Purpose: To evaluate various brain radiotherapy fractionation regimens by calculating relative damaged volume (RDV) of normal brain while giving the same biologically effective dose (BED) to tumor. Methods: The RDV was computed for 5 brain cases with the tumor sizes varying between 0.9–63.2 cm3 for 17 different fractionation regimens (1–8, 10, 12, 15, 20, 25, 30, 35, 40, and 50 fractions). A 20Gy single fraction regimen was used as the tumor control probability reference. The LQ model with correction for tumor regrowth effect (Tpot) was used to determine BED to tumor. The RDV was calculated from patients brain DVHs and a logistic local response function, which is characterized by DL50, the dose required to produce 50% local damage. Various Tpot (5 –30 days) and DL50 (20 – 50Gy) values were used to calculate RDV for each patient. Assuming a homogeneous brain structure, we used RDV as a surrogate for brain normal tissue complication probability (NTCP). A smaller RDV value represents lower NTCP and higher therapeutic ratio. Results: The minimum RDV among 17 studied fractionation regimens varied with Tpot, DL50, and tumor size. Stereotactic radiosurgery (SRS) is the favorable treatment regimen for low DL50 (20Gy). For a small sized tumor (

7522 Background: We have found that FDG-PET response during chemoradiation for patients with NSCLC is heterogeneous and predicts outcome. We hypothesized that dose escalated treatment targeted to the FDG-avid tumor would improve local tumor control. Methods: This is a phase II trial for patients with locally advanced, inoperable/unresectable NSCLC. Conformal radiotherapy (RT) was given in 30 daily fractions. RT dose was individualized to a fixed risk of lung toxicity and adaptively escalated to the residual tumor on mid-tx FDG-PET up to a total physical dose of 86 Gy. Patients had concurrent weekly followed by consolidation carboplatin/paclitaxel. The primary endpoint was local-regional tumor control (LRTC) at 2 years. Survival was calculated from RT start. Results were compared to stage-matched patients treated during the same time period with standard RT dosing (60-66 Gy). The data are presented as median (95% CI) unless otherwise specified. Results: 42 patients were enrolled: median age 63 years (range 45-83); 28 (67%) male; 39 (93%) smokers; 39 (93%) stage III; and 45% squamous cell. The mean gross tumor volume was 154 cc (range 10-617 cc). Median physical dose reached was 84 Gy (range 63-86 Gy), equivalent to 90 Gy in 2 Gy fractions (biological effective dose 108 Gy). 8 patients (19%) had RT-induced lung toxicity and 13 (31%) grade ≥2 esophagitis. Minimum and median follow-up were 10 and 25 months, respectively. The 2-year rates of in-field LRTC, overall LRTC, and LR-PFS were 84% (63-94%), 68% (47-82%), and 43% (27-58%), respectively. 14 patients progressed: 7 (50%) first at distant sites; 5 (36%) at nodal regions; 2 (14%) at primary tumor. Median overall survival was 26 months and 2-year overall survival rate was 51% (34-65%). These results compared favorably to stage-matched patients treated with standard-dose RT [2-year overall survival 23% (8-41%)]. Conclusions: These results support our hypothesis that adapting RT by escalating dose to the FDG avid region detected mid-tx improves 2-year local-regional tumor control. Adaptive RT may also improve overall survival. RTOG 1106 is currently testing this regimen in a randomized fashion. Clinical trial information: NCT01190527.

7580 Background: There is a growing literature on unique profiles of serum micro RNAs (miRNAs) expression to predict clinical outcome of metastatic and early stage non-small cell lung cancer (NSCLC). However, the predictive role of circulating miRNAs in unresectable NSCLC treated with definitive radiation therapy (RT) is unknown. Methods: 134 patients with inoperable/unresectable NSCLC treated with definitive RT (18-month minimum follow-up) were eligible. Serum samples were collected prospectively before treatment. 100 patients had enough serum and reliable miRNA profile quality, which were randomly divided into training and validation sets (50 patients each). MiRNA profiling was performed using real-time PCR-based array, containing a panel of 84 miRNAs detectable in human bodily fluids. Spiked-in cel-miR-39 was used for normalization. Stepwise regression Cox model building was used to build a miRNA signature on the training set, which was then assessed on the validation set both alone and with clinical factors. Results: The median age was 67 years; 76% were stages III and 79% received chemoradiation; the median physical dose was 70.0 Gy. A serum hsa-miR-885/hsa-miR-7 signature was identified as significant predictors for overall survival (OS) in the training set, which was validated by the validation set (p=0.02). After adjustment for GTV Volume and KPS, the only two significant clinical factors in univariate analysis, this signature remained significant (p=0.04). In the high-dose RT group (>70 Gy, n=45), individuals with low-risk had a significantly longer OS than patients with high-risk (70.7 vs. 18.8 months, p=0.007); while in the low-dose RT group (≤70 Gy, n=55), no significant association was observed (OS, 22.0 vs. 13.3 months, p=0.43). Conclusions: Circulating hsa-miR-885/hsa-miR-7 signature may be used as a putative non-invasive biomarker for predicting survival and radiation resistance in unresectable NSCLC, which may potentially help to select patients who will not benefit from high-dose radiation. Independent validation studies are needed to confirm our findings. Clinical trial information: NCT01190527.

7579 Background: SUVmaxat baseline FDG-PET has been reported as a significant prognostic factor while recent studies suggest that metabolic tumor volume (MTV) may be more important factor in patients with NSCLC. We hypothesized that TLG is a better prognostic factor than either SUVmax or MTV alone for overall survival (OS) and progression free survival (PFS) in NSCLC because it integrates both volumetric and biologic activity. Methods: The study population included a prospectively recruited cohort of stage I-III NSCLC patients treated with chemoradiation. FDG PET/CT scans were performed within 2 weeks from treatment start. The SUV in the tumor was normalized to that of the background level in the middle of ascending aorta to minimize the confounding effect from inter-scan variation in SUV measurement. MTV was delineated by auto-threshold at 1.5 times background level in the aorta followed by knowledge based manual editing. Mean and maximum SUV normalized to the background level were computed. TLG was calculated as the product of lesion SUVmean and MTV. Results: A total of 96 patients with minimum follow-up of 1 year were eligible. The median follow-up among survivors was 30 months. Univariate analysis demonstrated that MTV and TLG were significant factors for both OS and PFS (all Pmean and PFS (P=0.013), but there was no significant association between SUVmean and OS. SUVmax was not a significant factor for either OS or PFS (all P>0.05). Under multivariate Cox regression analysis, MTV (HR= 2.62, P= 0.003) and NSUVmean (HR=0.351, P=0.003) were significantly associated with PFS; but only TLG was significantly associated with OS (HR=2.14, P=0.006)adjusted by of TNM stage and other clinical factors. Conclusions: These results support our hypothesis that metabolic tumor volume and biologic average glucose metabolic activity of this volume are more important prognostic factors for overall prognosis than SUVmax in NSCLC patients treated with chemoradiation. Should this be validated by independent studies, future clinical trial should take this into consideration for individualized care.

Purpose: To determine the frequency with which elevated plasma transforming growth factor-β1 (TGFβ1) concentrations occur in lung cancer patients, to determine the kinetics of TGFβ1 expression during and after radiotherapy and to correlate plasma TGFβ1 levels with disease status after treatment. Materials and methods: Plasma samples were obtained before, during and after radiotherapy in 54 patients with lung cancer and 20 normal controls. Plasma TGFβ1 levels were measured using an enzyme-linked immunosorbent assay. Results: Baseline TGFβ1 levels in lung cancer patients and normal controls were 13.0 ± 2.5 and 4.4 ± 0.3 ng/ml, respectively. Elevated TGFβ1 were found in 50% ( 27 54 ) of lung cancer patients. During radiation therapy plasma TGFβ1 levels declined, however, by the completion of treatment the mean TGFβ1 level had not normalized in patients with lung cancer. The TGFβ1 level at last follow-up correlated with disease status in those patients with an increased pretreatment plasma level. Three of four patients with no evidence of cancer had normal follow-up TGFβ1 levels, compared to 2 16 patients with residual or recurrent tumor (P = 0.02). Conclusions: Elevated plasma TGFβ1 levels occur frequently in patients with lung cancer. In those patients with an elevated plasma TGFβ1 level at diagnosis, monitoring this level may be useful in detecting both disease persistence and recurrence after therapy.

OBJECTIVE: The authors examined the expression of transforming growth factor-beta receptor (TGF-beta r) types I and II and the mannose 6-phosphate/insulin-like growth factor-II receptor (M6-P/IGF-IIr) in human hepatocellular carcinoma (HCC). SUMMARY BACKGROUND DATA: Transforming growth factor-beta (TGF-beta) is part of a superfamily of peptide-signaling molecules that play an important role in modulating cell growth. It is secreted as a latent complex and therefore, must be activated to elicit a biological response. Bioactivation of the TGF-beta complex is facilitated by binding to the M6-P/IGF-IIr. Once activated, TGF-beta exerts its effects by binding to specific cell membrane TGF-beta receptors. The loss of responsiveness of hepatocytes to TGF-beta has been implicated in hepatocarcinogenesis and could result from a loss in the expression of either the TGF-beta receptors or the M6-P/IGF-IIr. METHODS: Human hepatocellular carcinomas and surrounding normal tissue were collected from operating room samples and snap-frozen in liquid nitrogen (n = 13). Tissues from two tumors were fixed in Omni-fix for sectioning and immunohistochemistry staining for the M6-P/IGF-IIr and TGF-beta 1. RNA was extracted from both normal and malignant liver tissue and analyzed using an RNase protection assay. SDS-PAGE of purified membrane hybridized with 125I-TGF-beta 1 and 125I-IGF-II was used to determine the TGF-beta type I (TGF-betarI) and type II (TGF-beta rII) receptors and M6-P/IGF-IIr protein levels, respectively. Gels were quantitated by phosphorimager, and a paired t test was used for statistical analysis. RESULTS: In HCC, a 60% (p < 0.01) and 49% (p < 0.02) reduction in the mRNA levels for T beta rI and T beta rII, respectively, relative to the receptor levels in surrounding normal liver, was shown. A similar decrease in the receptor protein levels also was observed. The M6-P/IGF-IIr mRNA and protein levels were reduced in 7 of 11 hepatocellular carcinomas. Immunohistochemical staining demonstrated an absence of intracellular TGF-beta 1 and reduced M6-P/IGF-IIr in the hepatocellular carcinoma cells. CONCLUSIONS: These results demonstrate that human HCCs have a significantly reduced expression of both the TGF-beta rI- and TGF-beta rII-signaling receptors for TGF-beta. This may provide a selective growth advantage to the HCC by allowing them to escape the mito-inhibitory effects of activated TGF-beta. Furthermore, in the subset of HCC in which the expression of the M6-P/IGF-IIr is downregulated, the bioactivation of TGF-beta also may be impaired.

e21115 Background: Radiation induced thoracic toxicities (RITT) in lung, esophagus and pericardium are dose-limited toxicities in patients with non-small cell lung cancer (NSCLC). Plasma transforming growth factor-beta1 (TGFβ1) and its change during radiotherapy was reported as a biomarker and single nucleotide polymorphisms (SNP) in TGFβ1 pathway was associated with RITT in our previous studies. In order to explore the mechanism of RITT, we tested whether plasma TGFβ1 was associated with genotypes of TGFβ1, tissue plasminogen activator (tPA) and angiotensin converting enzyme (ACE). Methods: Patients with stage I-III NSCLC enrolled in prospective clinical trials were eligible. All patients received radiotherapy with or without concurrent chemotherapy. Platelet-poor plasma was obtained pre-RT and at 4–5 weeks (40–50 Gy) during RT. Plasma TGF-β1 was measured using an enzyme-linked immunosorbent assay. The DNA samples extracted from blood before treatment were analyzed for the following genetic variations: TGFβ1 509C/T, TPA -7351 C/T, and ACE I/D. Results: 76 NSCLC patients received definitive radiotherapy (median dose 66 Gy) were enrolled. For the entire group of patients, the mean pre-RT TGFß1 level was 10.7±2.3 ng/ml, and the mean during-RT TGFß1 level was 6.0±0.7 ng/ml. No significant TGFß1 level differences were found at pre-RT and during-RT in patients with different genotypes of TGFß1 and tPA. Only ACE DD group had marginally higher pre-RT TGFß1 level than II and ID group (DD 23.6ng/ml vs. II 9.0ng/ml vs. ID 7.5ng/ml, p = 0.05), but no difference at during-RT(p=0.346) or during-RT/pre-RT ratio(p = 0.433). However, patients with TGFß1 509CC had higher elevation of plasma TGF ß1 level at fourth week during-RT than T allele carriers (TGFß1 level ratio of during-RT/pre-RT were CC 1.4±0.2 vs. T allele carriers 0.7±0.1, p=0.047). Conclusions: This exploratory study demonstrated that patients with TGFß1 509CC had higher elevation of plasma TGF ß1 level at fourth week during-RT than T allele carriers. This can help explain the correlation of TGF ß1 level elevation and higher risk of radiation induced lung toxicity in patients with NSCLC.

Purpose: To perform a systematic comparison of the Monte Carlo(MC), convolution/superposition (CS), and equivalent path length (EPL)‐based dose calculation algorithms for the purposes of outcomes modeling in lungcancertreatment planning.Methods: Several treatment plans (originally planned using EPL) from a large database of patients treated on a lungdose escalation protocol were retrospectively recalculated using MC and CS. Doses were computed in the homogeneous (unit‐density) and heterogeneous geometries; homogeneous calculations were used to elicit differences in the beam models To evaluate algorithmic differences due to heterogeneity effects, beam model differences were minimized by adjusting beam weights in the homogeneous plans to achieve the same prescribed dose with each algorithm. These beam weights were then applied to the heterogeneous geometries. Absolute dose distributions were compared using: color‐wash dose difference displays, isodose lines, EUD (for the target) and mean lungdose (MLD) and NTCP (for the normal lungs).Results: For the target, MC and CS‐computed EUDs were in good agreement for both homogeneous and heterogeneous cases, with maximum dose differences of 1.2 Gy noted. Differences between EPL and MC (or CS) were generally much larger, in the heterogeneous plans extending up to 6 Gy. Differences in MLD computed with MC and CS ranged between 2% and 15% in the heterogeneous plans. These differences were similar in the corresponding homogeneous geometries, illustrating the importance of beam model disparities. For EPL, differences in the MLD and NTCP (relative to MC or CS) were much larger in the heterogeneous plans indicating systematic differences in the normal lungdose prediction. Conclusion: Evidence thus far is suggestive that discrepancies in dose computed with EPL and MC (or CS) will lead to differences in correlations of dose with outcome with respect to the target as well as normal tissue complications (radiation induced pneumonitis) and calculated NTCP.

Purpose To study the pattern of local failure for patients of non-small cell lung cancer (NSCLC) treated on a conformal therapy dose escalation protocol. Materials/Methods Subjects included patients who had failed locally and a matched group without failures after 3-D conformal radiation per a phase I dose escalation trial. Radiation doses were computed using an equivalent path length algorithm, ranged from 65.1 to 102.9 Gy in 2.1 Gy fractions. The original and recurrent gross target volume (GTV and RGTVs) were contoured by a radiologist and a radiation oncologist. The original and planning target volume (PTVs and RPTVs) were from GTVs plus 1-1.5 cm expansion. DVHs of each structure were constructed and generalized equivalent uniform doses (EUD = {Σi (di)a}1/a) were also computed. Marginal failure was defined for RGTVs covered by the original 10% to 90% isodose surfaces, while central failures were inside the original 90-100% isodose surfaces. Results There are no significant differences between the failed and controlled groups with regard to average original GTV volumes, average total prescription doses, mean GTV doses, or mean PTV doses, and minimal PTV doses. Of the 15 RGTVs, five are located across the margins of the original PTVs, nine mostly within the PTVs, and one outside of the original PTVs. The average EUDs for the RGTVs and RPTVs are 60 ± 30 and 49 ± 31 Gy, respectively, significantly below the prescribed doses (p = 0.03). Forty percent RGTVs and 60% RPTVs had EUDs less than 66 Gy (a common prescription dose). The RGTVs are not encompassed by the original 90% isodose surfaces for 7 of the 15 cases, all of which failed at the margins of or outside the original PTVs. The EUDs for these seven RGTVs are much less than (1%-63% of) the EUDs for the original PTVs. When the doses are recalculated using the Monte Carlo algorithm, additional 10-20% dose reductions are found for patients who failed after high dose radiation. Conclusions Inadequate dose, marginal miss and under-dose from inaccurate calculation algorithm all contributed to the local failure after 3-D conformal radiation therapy. In addition to dose escalation, further clinical trials should focus on margin definition and accurate dose calculation to improve the tumor control in NSCLC.

These data demonstrate that plasma TGFBl is a independent marker which could be used to monitor the persistence and/or recurrence of disease in lung cancer patients. The end RT TGFDl levels might be used to choose patients who need further adjuvent treatment.

Superior vena cava syndrome (SVCS) is a relatively common sequela of mediastinal malignancies and may cause significant patient distress. SVCS is a medical emergency if associated with laryngeal or cerebral edema. The etiologies and management of SVCS have evolved over time. Non-malignant SVCS is typically caused by infectious etiologies or by thrombus in the superior vena cava and can be managed with antibiotics or anti-coagulation therapy, respectively. Radiation therapy (RT) has long been a mainstay of treatment of malignant SVCS. Chemotherapy has also been used to manage SVCS. In the past 20 years, percutaneous stenting of the superior vena cava has emerged as a viable option for SVCS symptom palliation. RT and chemotherapy are still the only modalities that can provide curative treatment for underlying malignant etiologies of SVCS. The first experiences with treating SVCS with RT were reported in the 1970’s, and several advances in RT delivery have subsequently occurred. Hypo-fractionated RT has the potential to be a more convenient therapy for patients and may provide equal or superior control of underlying malignancies. RT may be combined with stenting and/or chemotherapy to provide both immediate symptom palliation and long-term disease control. Clinicians should tailor therapy on a case-by-case basis. Multi-disciplinary care will maximize treatment expediency and efficacy.

Masses in the anterior mediastinum can be neoplasms (eg, thymomas, thymic carcinomas, or lung metastases) or non-neoplastic conditions (eg, intrathoracic goiter). Thymomas are the most common primary tumor in the anterior mediastinum, although they are rare. Thymic carcinomas are very rare. Thymomas and thymic carcinomas originate in the thymus. Although thymomas can spread locally, they are much less invasive than thymic carcinomas. Patients with thymomas have 5-year survival rates of approximately 78%. However, 5-year survival rates for thymic carcinomas are only approximately 40%. These guidelines outline the evaluation, treatment, and management of these mediastinal tumors.

e17530 Background: The transforming growth factor beta (TGFβ) pathway, an important regulator in cellular metabolic process, has been reported for significant association with cancer prognosis. This study was to exam the association between single nucleotide polymorphisms (SNPs) of TGFβ pathway and overall survival (OS) in subjects with non-small cell lung cancer (NSCLC). Methods: Patients with stage I-III NSCLC received definitive radiotherapy with/without chemotherapy were eligible to this prospective study. The primary endpoint was OS which was calculated from radiation treatment start to death or censored. DNA samples for genotyping were extracted from buffy-coat which was collected before commencement of treatment. 19 SNPs in 10 genes (BMP1, BMP2, INHBC, SMAD1, SMAD3, SMAD4, SMAD6, SMAD7, SMAD8, TGFβ1), which was reported to have significant correlation with OS of lung cancer, were selected. MassArray System (Sequenom Company) was used for genotyping. Cox regression was performed for multivariate analysis to examine the effects of genotypes on OS using dominant and recessive genetic model. Results: 126 consecutive patients, 91% of them were Caucasian, were recruited in this study. All SNPs call rates were over 90%. Assay reproducibility was over 99% by random double-blinding duplicate or triplicate genotyping. Among clinical factors analyzed, radiation dose was only significant independent factor predicting OS (P=0.001). Genotypic association study showed that 7 SNPs (rs235756, rs11939979, rs12102171, rs6494633, rs12456284, rs12906898 and rs4803455) were significantly associated with OS, adjusted for age, gender, smoking, histology, clinical stage, tumor volume, Karnofsky Performance Status, radiotherapy dose, and chemotherapy. The strongest association was in SMAD3: rs12102171 (P=0.004, HR=2.28, 95%CI, 1.26-4.15). Conclusions: This study partly validated findings from previous studies that genetic variations in the TGFβ pathway are significant predictors of overall survival in NSCLC patients treated with definitive radiotherapy with/without chemotherapy.