Purpose/Objective(s)The incidence of human papillomavirus-positive (HPV+) oropharyngeal cancer (OPC) has increased rapidly in North America. HPV serology has been proposed as a sensitive, scalable, and cost-effective early detection test. HPV seropositivity can precede clinical presentation of OPC by several years, so additional diagnostic and/or surveillance procedures may be necessary to optimize early cancer detection. The potential for HPV circulating tumor DNA (ctDNA) to confirm a diagnosis of OPC in seropositive individuals is poorly understood. Moreover, the concordance of baseline HPV serology and HPV ctDNA and their respective associations with cancer burden in OPC patients are not known. We hypothesized that plasma levels of HPV antibodies and HPV ctDNA are associated with disease burden in a cohort of p16+ OPC patients.Materials/MethodsWe analyzed pre-treatment peripheral blood plasma from 100 patients with non-metastatic p16+ OPC treated with definitive (chemo)radiotherapy. A multiplex ELISA was used for serologic evidence of HPV proteins (L1, E1, E2, E4, E6, E6) from 10 HPV genotypes, quantified by mean fluorescence intensity (MFI). Plasma HPV ctDNA was quantified by whole viral genome sequencing utilizing a custom HPV-targeted capture panel for 38 HPV genotypes (HPV-seq). Gross tumor volume (GTV) was obtained from the sum of all target contours on computed tomography simulation scans. Statistical tests Kruskal-Wallis and Pearson's r were used to compare the assays.ResultsSeropositivity and HPV ctDNA positivity were found in 100% of cases. Serological results indicated 10 genotypes, and HPV ctDNA results indicated 9 distinct genotypes; both assays identified HPV16 as the most prevalent genotype (98% for each). Among the 98 patients with detectable HPV16 within ctDNA, HPV16 E6 seropositivity demonstrated high sensitivity (95%), while E1 (50%), E4 (47%), E7 (66%), and L1 (63%) demonstrated considerably lower sensitivities. HPV ctDNA but not HPV16 E6 MFI was positively associated with disease burden as determined by T- and N-category (Table 1), and by GTV (ctDNA vs GTV, r=0.47 p=1.0e-06; E6 MFI vs GTV, r=-0.14 p=0.17).ConclusionThis is the first report to compare HPV serology and HPV ctDNA sequencing in OPC patient plasma. Both tests were highly sensitive and could identify multiple oncogenic HPV genotypes. The level of HPV16 ctDNA but not HPV16 E6 antibodies in plasma was associated with disease burden. These findings have potential implications for early detection and prognostication.

Unscheduled treatment interruptions may lead to reduced quality of care in radiation therapy (RT). Identifying the RT prescription dose effects on the outcome of treatment interruptions, mediated through doses distributed into different organs-at-risk (OARs), can inform future treatment planning. The radiation exposure to OARs can be summarized by a matrix of dose-volume histograms (DVH) for each patient. Although various methods for high-dimensional mediation analysis have been proposed recently, few studies investigated how matrix-valued data can be treated as mediators. In this paper, we propose a novel Bayesian joint mediation model for high-dimensional matrix-valued mediators. In this joint model, latent features are extracted from the matrix-valued data through an adaptation of probabilistic multilinear principal components analysis (MPCA), retaining the inherent matrix structure. We derive and implement a Gibbs sampling algorithm to jointly estimate all model parameters, and introduce a Varimax rotation method to identify active indicators of mediation among the matrix-valued data. Our simulation study finds that the proposed joint model has higher efficiency in estimating causal decomposition effects compared to an alternative two-step method, and demonstrates that the mediation effects can be identified and visualized in the matrix form. We apply the method to study the effect of prescription dose on treatment interruptions in anal canal cancer patients.

Purpose/Objective(s)Plasma HPV DNA could detect molecular residual disease (MRD) after (chemo)radiotherapy (CRT/RT) for oropharyngeal cancer (OPC). Whole HPV genome sequencing (HPV-seq) could both genotype and quantify HPV DNA more sensitively than digital polymerase chain reaction (dPCR). We compared HPV-seq and dPCR pre- and post-RT in a large, prospective OPC cohort. We hypothesized that HPV-seq is more sensitive and more accurately predicts recurrence by MRD.Materials/MethodsWe recruited patients (pts) with non-metastatic p16-positive OPC (TNM8 stage I-III) undergoing curative-intent CRT/RT. Each pt had peripheral blood plasma prospectively collected at pre-RT and 3-months post-RT. The primary endpoint was recurrence-free survival (RFS) and secondary endpoint was overall survival (OS). Time-to-event analysis was performed using the Kaplan-Meier method with log-rank tests and multivariable Cox regression. HPV-seq was performed using a validated hybrid capture panel targeting 38 HPV genomes and 15 recurrently mutated genes in OPC. dPCR was performed using a validated multiplexed assay targeting HPV16 E6/E7. HPV DNA levels were expressed as copies/mL plasma and log-transformed for statistical tests.ResultsWe performed HPV-seq and dPCR on pre- and post-RT samples from 245 pts. Median follow-up was 52.3 (4.5-95.8) months, and 3-yr RFS was 87.9%. HPV-seq detected plasma HPV DNA in 237 (97%) pts pre-RT (HPV16-only: 161, non-HPV16: 16, mixed: 60) with 0.017-364867 copies/mL. Of pre-RT samples with HPV-16 reads detected on HPV-seq, HPV-16 dPCR was detectable in 202/221 (91%), with 0.32-37313 copies/mL on dPCR. Copies of HPV16 DNA per mL were strongly correlated between HPV-seq and dPCR (R=0.986). Post-RT, HPV-seq detected residual HPV DNA in 48 (20%) pts. Of these, dPCR detected 19 and did not detect 29, of which 6 had non-HPV16, and the other 23 had HPV16 with <1.37 copies/mL. Of pts with 3-year recurrence, post-RT detection of HPV DNA by HPV-seq predicted 16/23 (sensitivity=70%, NPV=95.5%, AUC=0.81) and by dPCR predicted 7/23 (sensitivity=30%, NPV=91.0%, AUC=0.63). Post-RT residual HPV DNA on HPV-seq was associated with poorer RFS (3-yr 58% vs. 96%, p<0.0001) and OS (3-yr 61.4% vs. 97.1%, p<0.0001). Of 192 pts with HPV16 undetectable by dPCR at post-RT, the 23 with residual HPV DNA detected by HPV-seq still had poorer RFS (3-yr 62.5% vs. 95.7%, p<0.0001) and OS (3-yr 75.0% vs. 96.9%, p<0.0001) than those with HPV-seq clearance. Adjusting for T and N staging, post-RT residual HPV16 DNA was independently prognostic of RFS and OS by either assay, with odds ratios for RFS of 14.91 [6.19, 35.94] for HPV-seq and 5.17 [2.09, 12.81] for dPCR.ConclusionWhole viral genome sequencing genotypes and quantifies plasma HPV DNA. Pts with residual HPV DNA had poorer RFS and OS, independent of stage. For HPV16, HPV-seq and dPCR are strongly correlated, but HPV-seq is more sensitive with higher negative predictive value and predicted 2.3x more 3-year recurrences in OPC pts.

Simple 3D architectures consistently outcompete more complex networks by quantitative measures. Qualitative assessment for clinical acceptability may not agree with quantitative performance, especially when the entire range of OARs is evaluated.

The term oligometastatic (OM) was introduced to describe patients with a limited number of clinically detectable metastases for which local ablative therapy could potentially provide improved disease-free survival or even cure. The incidence of an oligometastatic disease distribution is difficult to estimate. The purpose of this analysis was to describe the reasons precluding trial participation in our trial of SBRT for oligometastatic cancer. We retrospectively audited all patients who were screened for participation in a single-arm phase II trial designed for patients with OM cancer (up to 5 sites of extra-cranial metastases). Study participants received local ablative therapy to all known sites of disease and at least one index lesion was treated with 5 fraction SBRT (study intervention). Eligibility criteria required baseline computed tomography (CT) ≤1 month prior to enrolment, and all cases were discussed at peer-review rounds to confirm eligibility. Between Mar 7 2013 and Nov 17 2017, 236 patients were screened and 145 (61%) enrolled to the study. The reasons for screen failure included patients declining study participation in 4% (9/236) [n = 9; 4 with <5 mets, and 5 with uncertain number] and 12% (29/236) having >5 metastases upon restaging baseline scans. Of the remaining 53 patients who had ≤5 metastases but were not enrolled, reasons for screen failure included uncertainty that the primary tumor was controlled (n=9), lesions where alternative ablative therapy was recommended (n=10), not suitable for SBRT (n=18; size n=3, non-SBRT strategy preferred n=8, toxicity risk n=4, previous RT field overlap n=3), high subjective risk of systemic relapse (n=9; progressive disease during restaging scans n=4, not suitable for comprehensive ablative therapy n=1, plan for systemic therapy n=4) or other reasons (n=7; e.g. index lesion resolution, index lesion was a primary tumor, presence of cranial metastases). In our review of a cohort of patients flagged to be potential candidates for enrollment in a clinical trial of SBRT for OM cancer, 87% of patients were confirmed to have <5 metastases. The screen failure rate was high. The most common reason for exclusion was detection of >5 metastases (12%) while a small group of patients (5%) were deemed to be at high risk of systemic relapse based on clinical factors and peer review rounds. Reporting of the characteristics of screen failed patients are important in the interpretation of trial results.

PurposeHigh-dose-rate (HDR) intravaginal brachytherapy is an important adjuvant treatment for endometrial cancer and is typically delivered using vaginal cylinder applicators with fixed diameters. Best practice recommends fitting patients with the largest diameter cylinder that can be comfortably accommodated. Commercial vendors currently offer four standard diameter sizes: XS (2.0 cm), S (2.5 cm), M (3.0 cm), and L (3.5 cm). While the range of these diameters encompasses the sizing needs of the population generally, in our experience, there is significant clinical need for a cylinder option in between the available sizes 2.0 cm and 2.5 cm. We aimed to highlight a clinical need for this applicator by analyzing the demographics of the patient population at our institution who benefited from treatment with cylinder diameters smaller than the commercially available 2.5 cm size.Materials and MethodsUsing a prospective database approved by the Institutional Review Board, we reviewed the records of 375 consecutive stage I-III endometrial patients who received vaginal HDR brachytherapy as part of adjuvant treatment between 09/2011 and 12/2015. Each patient was treated with the largest size comfortably fitted vaginal cylinder, with custom-fabricated diameter sizes of (cm) 2.0, 2.3, 2.6, and 3.0. The need for a cylinder smaller than 2.5 cm (i.e. 2.3 cm or less in our cohort) was the outcome variable of interest. The Chi-squared test was used to assess the association of small cylinder requirement with dichotomous variables of interest including age, body mass index (BMI), race, gravidity (G), stage, grade, histology, and receipt of chemotherapy/radiotherapy. Multivariable logistic regression analysis was performed to calculate adjusted odds ratios (ORs) for independent predictors of need for cylinder smaller than 2.5 cm.ResultsOverall, cylinders smaller than 2.5 cm were used in 60 patients (16%) out of the overall cohort (57 with size 2.3 cm and 3 with size 2.0 cm). On univariate analysis, normal BMI vs. overweight or obese (p <.001), G0 vs. ≥ G1 (p <.001), and Caucasian race vs. minority (p =.034) were predictors of need for cylinder smaller than 2.5 cm. On multivariable logistic regression, G0 vs. ≥ G1 status (OR 5.03, 95% CI 2.55 - 9.90, p <.001) and normal BMI vs. overweight or obese (OR 3.06, 95% CI 1.53 - 6.12, p =.002) remained significant predictors of need for cylinder smaller than 2.5 cm. Among the subgroup of women who were G0 with normal BMI, 63.2% (12/19) required cylinder size 2.3 cm whereas none required 2.0 cm.ConclusionsTable 1Patient characteristics by vaginal cylinder size (n = 375)2.0 or 2.3 cm cylinder2.6 or 3.0 cm cylinderUnivariable(n = 60)(n = 315)p-valueAge.203 <6532 (18.8%)138 (81.2%) ≥6528 (13.7%)177 (86.3%)BMI<.001 Normal23 (35.9%)41 (64.1%) Overweight or obese37 (11.9%)274 (88.1%)Race.034 White56 (17.8%)258 (82.2%) Minority4 (6.6 %)57 (93.4%)Gravidity<.001 GO27 (39.7%)41 (60.3%) ≥G133 (10.9%)269 (89.1%)Stage.265 I-II53 (17.0%)258 (83.0%) IIII7 (10.9%)57 (89.1%)Grade.387 1-235 (17.2%)169 (82.8%) 322 (13.7%)139 (86.3%)Histology.536 Endometrioid45 (16.9%)221 (83.1%) High risk15 (13.8%)94 (86.2%)Chemotherapy.325 Yes23 (13.9%)143 (86.1%) No37 (17.7%)172 (82.3%)EBRT.758 Yes4 (19.0%)17 (81.0%) No56 (15.8%)298 (84.2%)EBRT = external beam radiotherapy; BMI = body mass index Open table in a new tab PurposeHigh-dose-rate (HDR) intravaginal brachytherapy is an important adjuvant treatment for endometrial cancer and is typically delivered using vaginal cylinder applicators with fixed diameters. Best practice recommends fitting patients with the largest diameter cylinder that can be comfortably accommodated. Commercial vendors currently offer four standard diameter sizes: XS (2.0 cm), S (2.5 cm), M (3.0 cm), and L (3.5 cm). While the range of these diameters encompasses the sizing needs of the population generally, in our experience, there is significant clinical need for a cylinder option in between the available sizes 2.0 cm and 2.5 cm. We aimed to highlight a clinical need for this applicator by analyzing the demographics of the patient population at our institution who benefited from treatment with cylinder diameters smaller than the commercially available 2.5 cm size. High-dose-rate (HDR) intravaginal brachytherapy is an important adjuvant treatment for endometrial cancer and is typically delivered using vaginal cylinder applicators with fixed diameters. Best practice recommends fitting patients with the largest diameter cylinder that can be comfortably accommodated. Commercial vendors currently offer four standard diameter sizes: XS (2.0 cm), S (2.5 cm), M (3.0 cm), and L (3.5 cm). While the range of these diameters encompasses the sizing needs of the population generally, in our experience, there is significant clinical need for a cylinder option in between the available sizes 2.0 cm and 2.5 cm. We aimed to highlight a clinical need for this applicator by analyzing the demographics of the patient population at our institution who benefited from treatment with cylinder diameters smaller than the commercially available 2.5 cm size. Materials and MethodsUsing a prospective database approved by the Institutional Review Board, we reviewed the records of 375 consecutive stage I-III endometrial patients who received vaginal HDR brachytherapy as part of adjuvant treatment between 09/2011 and 12/2015. Each patient was treated with the largest size comfortably fitted vaginal cylinder, with custom-fabricated diameter sizes of (cm) 2.0, 2.3, 2.6, and 3.0. The need for a cylinder smaller than 2.5 cm (i.e. 2.3 cm or less in our cohort) was the outcome variable of interest. The Chi-squared test was used to assess the association of small cylinder requirement with dichotomous variables of interest including age, body mass index (BMI), race, gravidity (G), stage, grade, histology, and receipt of chemotherapy/radiotherapy. Multivariable logistic regression analysis was performed to calculate adjusted odds ratios (ORs) for independent predictors of need for cylinder smaller than 2.5 cm. Using a prospective database approved by the Institutional Review Board, we reviewed the records of 375 consecutive stage I-III endometrial patients who received vaginal HDR brachytherapy as part of adjuvant treatment between 09/2011 and 12/2015. Each patient was treated with the largest size comfortably fitted vaginal cylinder, with custom-fabricated diameter sizes of (cm) 2.0, 2.3, 2.6, and 3.0. The need for a cylinder smaller than 2.5 cm (i.e. 2.3 cm or less in our cohort) was the outcome variable of interest. The Chi-squared test was used to assess the association of small cylinder requirement with dichotomous variables of interest including age, body mass index (BMI), race, gravidity (G), stage, grade, histology, and receipt of chemotherapy/radiotherapy. Multivariable logistic regression analysis was performed to calculate adjusted odds ratios (ORs) for independent predictors of need for cylinder smaller than 2.5 cm. ResultsOverall, cylinders smaller than 2.5 cm were used in 60 patients (16%) out of the overall cohort (57 with size 2.3 cm and 3 with size 2.0 cm). On univariate analysis, normal BMI vs. overweight or obese (p <.001), G0 vs. ≥ G1 (p <.001), and Caucasian race vs. minority (p =.034) were predictors of need for cylinder smaller than 2.5 cm. On multivariable logistic regression, G0 vs. ≥ G1 status (OR 5.03, 95% CI 2.55 - 9.90, p <.001) and normal BMI vs. overweight or obese (OR 3.06, 95% CI 1.53 - 6.12, p =.002) remained significant predictors of need for cylinder smaller than 2.5 cm. Among the subgroup of women who were G0 with normal BMI, 63.2% (12/19) required cylinder size 2.3 cm whereas none required 2.0 cm. Overall, cylinders smaller than 2.5 cm were used in 60 patients (16%) out of the overall cohort (57 with size 2.3 cm and 3 with size 2.0 cm). On univariate analysis, normal BMI vs. overweight or obese (p <.001), G0 vs. ≥ G1 (p <.001), and Caucasian race vs. minority (p =.034) were predictors of need for cylinder smaller than 2.5 cm. On multivariable logistic regression, G0 vs. ≥ G1 status (OR 5.03, 95% CI 2.55 - 9.90, p <.001) and normal BMI vs. overweight or obese (OR 3.06, 95% CI 1.53 - 6.12, p =.002) remained significant predictors of need for cylinder smaller than 2.5 cm. Among the subgroup of women who were G0 with normal BMI, 63.2% (12/19) required cylinder size 2.3 cm whereas none required 2.0 cm. ConclusionsTable 1Patient characteristics by vaginal cylinder size (n = 375)2.0 or 2.3 cm cylinder2.6 or 3.0 cm cylinderUnivariable(n = 60)(n = 315)p-valueAge.203 <6532 (18.8%)138 (81.2%) ≥6528 (13.7%)177 (86.3%)BMI<.001 Normal23 (35.9%)41 (64.1%) Overweight or obese37 (11.9%)274 (88.1%)Race.034 White56 (17.8%)258 (82.2%) Minority4 (6.6 %)57 (93.4%)Gravidity<.001 GO27 (39.7%)41 (60.3%) ≥G133 (10.9%)269 (89.1%)Stage.265 I-II53 (17.0%)258 (83.0%) IIII7 (10.9%)57 (89.1%)Grade.387 1-235 (17.2%)169 (82.8%) 322 (13.7%)139 (86.3%)Histology.536 Endometrioid45 (16.9%)221 (83.1%) High risk15 (13.8%)94 (86.2%)Chemotherapy.325 Yes23 (13.9%)143 (86.1%) No37 (17.7%)172 (82.3%)EBRT.758 Yes4 (19.0%)17 (81.0%) No56 (15.8%)298 (84.2%)EBRT = external beam radiotherapy; BMI = body mass index Open table in a new tab EBRT = external beam radiotherapy; BMI = body mass index

The trend in cancer therapy is toward individualized treatments based on patient-specific biology. However, the application of imaging procedures in the radiotherapeutic management of cancers remains nonpersonalized: a "one-protocol-fits-all" practice is often applied in the clinic worldwide. Essentially, the imaging protocols provided by manufacturers are uniformly applied without considering individual differences of patients being scanned. As such, children and petite adults may be overexposed from the default site-specific protocols (eg high-quality head or pelvis protocols) owing to reduced tissue attenuation.

Inflammatory Bowel Disease (IBD) including Crohn Disease (CD) and Ulcerative Colitis (UC) has traditionally been considered a relative contraindication to pelvic radiation therapy (RT). This retrospective study was conducted to investigate acute and late toxicity in patients with IBD receiving pelvic RT. IBD patients receiving curative or adjuvant doses of pelvic RT (BED2 greater than 75) between January 1, 1999 and September 30, 2012 were included. Abstracted data included diagnosis, treatment technique (conventional, 3D or IMRT), radiation dose, treatment compliance, and acute and late RTOG toxicity scores. Fisher’s Exact Test was used to assess relationship between patient characteristics and toxicity. The mean follow-up was 40.8 months (range: 4.7 to 159). Sixty-one patients were identified including 30 (49%) with CD, 28 (46%) with UC and 3 (5%) with IBD not otherwise specified. The most common cancer diagnoses were rectal (n = 17, 28%), anal (n = 18, 30%) and prostate (n = 17, 28%). Active IBD requiring medication was documented in 31/61 patients (51%). Acute ≥ grade 3 GI toxicity was observed in 7/61 patients (11%), all of whom had gastrointestinal malignancies treated with concurrent chemo-radiation therapy. Two patients (3%) died on treatment after being admitted for severe diarrhea. Overall, 6/61 patients (10%) were unable to complete RT (1-10 missed fractions, mean = 6). An additional 7/61 patients (12%) had unplanned RT break (1 to 24 days). On univariate analysis, gastrointestinal malignancy (p = 0.02) and concurrent chemotherapy (p = 0.05) significantly predicted for acute toxicity. IBD type, active IBD medication and previous IBD surgery was not predictive of acute toxicity. Following treatment, 10 patients (17%) with GU, GI and gynecologic malignancies had documented IBD flares and 3 patients (5%) had active IBD at last follow-up. Sixteen patients had rectal cancer treated with concurrent RT and chemotherapy with one late fistula (6%) and one late bowel obstruction (6%) documented in this group. Both patients had active IBD at diagnosis. The decision to treat IBD patients with pelvic radiation therapy is a difficult one but consideration is necessary for locally advanced, life- threatening cancers. The results of this large series investigating treatment-related toxicity suggest that pelvic radiation therapy should continue to be used with caution particularly for GI malignancies requiring pelvic RT and concurrent chemotherapy.

Purpose: Synthetic polymers have been used to coat the surface of existing low‐energy brachytherapy sources to improve their fixity in tissue in order to minimize seed migration in the patient. The purpose of this study was to evaluate the effects of the polymer‐coating on the photon energy spectra emitted by Pd‐103 seeds and to estimate its impact on the dosimetric characteristics of these seeds. Methods: Two Pd‐103 AnchorSeeds (Biocompatibles, Inc) and two Pd‐103 TheraSeeds (Theragenics Corporation) were used in this study. The AnchorSeed is made from the TheraSeed by coating its surface with bio‐absorbable polymeric anchoring rings and ribs. Photon energy spectra emitted by these seeds were measured using a high‐resolution, high‐purity Germanium (HPGe) photon spectrometer. The effects of polymer coating on the dose rate constant were quantified by comparisons of the coated seed with a non‐coated seed. Results: The relative photon energy spectrum emitted by the AnchorSeed was nearly identical to that emitted by the TheraSeed without polymer coating. The dose‐rate constants determined from the emitted photon energy spectra for the two seed models were identical to the third digit after the decimal point (0.679 ± 0.037 cGyh‐1U‐1). Because the basic dosimetric property of a brachytherapy seed is fundamentally determined by the photon characteristics emitted by the seed, effects of the polymer‐coating on the relative dosimetry data of TheraSeed is also expected to be minimal. Using a preexisting photon spectrometer, these measurements and calculations were completed in about 10 working hours. Conclusion: The polymer coating used in the AnchorSeed has minimal impact on its dosimetric characteristics compared to the non‐coated TheraSeed. We also demonstrate that photon spectroscopy provides a ready tool for the evaluation of impact of various coatings or other material changes in the seed manufacture process.

The ability of frequency-domain photothermal radiometry (PTR) and modulated luminescence (LUM) to detect secondary caries is presented. Signal behavior upon sequential demineralization and remineralization of a spot (diameter ~1 mm) on a vertical wall of sectioned tooth samples was investigated experimentally. From these studies, it was found that PTR–LUM signals change, showing a certain pattern upon progressive demineralization and remineralization. PTR amplitudes slightly decreased upon progressive demineralization and slightly increased upon subsequent remineralization. The PTR phase increased during both demineralization and remineralization. LUM amplitudes exhibit a decreasing trend at excitation/probe distances larger than 200 μm away from the edge for both demineralization and remineralization; however, at locations close to the edge (up to ~200 μm), LUM signals slightly decrease upon demineralization and slightly increase during subsequent remineralization.

To report a single-isocenter multi-segment (SiMs) conformal arc technique for efficient planning and delivery of SBRT treatments for lung cancer. The increasing use of SBRT for tumors in lung, liver and other extra-cranial sites has presented a growing challenge for treatment planning with traditional 3D non-coplanar beam techniques (NCBT) due to (1) potential patient-machine collision and (2) involuntary patient motion caused by frequent couch movements. A SiMs arc technique, applicable to tumors located on or off isocenter, has been developed to overcome these problems. The technique has since been used in our clinic for 15 lung SBRT patients. Patients were immobilized using whole-body VacLock bags and received 4D-CT scans. Planning volumes were contoured on 4D-CT movie loops and transferred to average intensity projection CTs for dose calculation and evaluation. Treatment plans were designed on the Eclipse treatment planning system using 6MV photons and AAA dose calculation algorithm. The SiMs arc technique involves: 1) identifying an isocenter, usually not in the tumor volume, that allows 360 degree free gantry rotation around the patient; 2) creating six isocentric dynamic conformal arc segments, each spanning about 60 degrees, from 1 to 359 degrees; the PTV-defined dynamic apertures of alternating arc segments were modified with unique margins superiorly and inferiorly to achieve a desired dose fall off in these directions, and 3) normalizing the plan so that the PTV was covered adequately. Dosimetric quality of the SiMs arc plans was compared to that of 3D NCBT plans. The use of SiMs dynamic conformal arc produced highly conformal dose coverage of tumor volume while keeping normal tissue doses within acceptable limits. Compared to 3D NCBT, the SiMs arc plans yielded conformal distributions which met both the requirements for tumor coverage and normal tissue sparing. Using a SiMs arc template plan, the time and effort needed to generate an acceptable SBRT plan was greatly reduced compared to 3D NCBT. For the same prescribed dose, the total treatment time was reduced by more than 30% compared to 3D NCBT. The use of SiMs arc technique eliminated need for inter-beam couch rotation and, hence, involuntary patient movements caused by sudden acceleration and deceleration of the treatment couch that could occur in 3D NCBT. A single-isocenter multi-segment conformal arc technique has been proposed, tested, and successfully applied to SBRT treatments for tumors in lung. Our experience demonstrated that this technique has not only improved the efficiency and dosimetric consistency of SBRT treatment planning but also led to more robust treatment delivery. The details of this technique for lung SBRT cases and its application to liver will be discussed.

15671 Background: This phase I study assessed the maximum tolerated dose (MTD) of 2-weekly cisplatin, infusional 5-FU and concurrent conformal radiotherapy (RT) for resected gastric cancer. Methods: Thirty four patients with resected stage IB to IV(M0) gastric adenocarcinoma were prescribed 12 weeks of infusional 5FU 200 mg/m2/day with RT 45Gy in 25 fractions starting at day 16. Cisplatin dose escalation occurred after a minimum of 3 patients without dose limiting toxicity (DLT) per cohort. DLT was defined as ≥ grade 3 toxicity requiring treatment or hospitalization; hematologic or biochemical changes not requiring treatment were not considered DLT. MTD was the highest dose level where no more than 1 of 6 patients had DLT. If 1 patient had DLT another 3 were treated at the same dose level. If ≥2 had DLT in any cohort that level exceeded MTD and the previous dose level was expanded to 6 patients. Results: Median age was 56 years (range 31–77); M:F 53:47%. Median follow up was 2.5 years (0.3–5). There were no grade 5 events. Five patients experienced DLT (see table). Cohort 5 had 2 patients with DLT exceeding MTD; cohort 4 was defined as the MTD. Acute toxicity included skin, weight loss, fatigue, anorexia, nausea, dehydration, hand-foot syndrome, sepsis, hyponatremia and hematological events. Eleven patients did not complete prescribed 5FU (cohort 1[1], 2[3], 3[3], 4[1], 5[3]) among whom the mean duration of 5FU received was 7 weeks (0.5–11). Of these 11 patients, 4 did not complete cisplatin (cohort 2[1], 5[3]); 3 did not complete RT on study receiving 0, 0 and 36Gy. One patient had late grade 3 toxicity: weight-loss and G-tube dependence. Nine patients relapsed of whom 6 died; 3-year overall and relapse free survival were 86% and 71% with median survival not reached. Conclusion: MTD occurred in cohort 4 with no DLT in this cohort. A phase II study is accruing patients using 12 weeks of infusional 5FU 200 mg/m2/day and 4 cycles cisplatin 40 mg/m2 2-weekly with RT 45Gy in 25 fractions starting at day 16. Acute and dose limiting toxicity in each cohort Cohort Cisplatin mg/m2 Cycles of cisplatin N Number of patients with toxicity Type of non hematological grade 3 toxicity+ All grade 1/2 Non-hematological grade 3 Hematological grade 3/4 1 0 0 6 6 1* 1 Sepsis* Hyponatremia 2 20 4 10 9 1* 0 Fatigue* 3 30 4 7 7 1* 4 Vomiting with dehyration* Stomatitis 4 40 4 6 6 2 0 Hyponatraemia not requiring treatment 5 40 5 5 5 2* 3 Anorexia* Dehyradtion* + Note that several patients experienced multiple non-hematologic and/or dose limiting toxicities; * Dose limiting toxicity No significant financial relationships to disclose.