Search

457 Background: Precision image-guided radiotherapy (RT) for patients (pts) with muscle-invasive bladder cancer (MIBC) is limited by daily anatomic variation and difficulty visualizing the tumor bed. We hypothesized that TraceIT, a radiopaque resorbable hydrogel, can be injected around the tumor bed to improve image guidance. We present the results of our pilot trial (NCT03125226) evaluating safety/feasibility of TraceIT in pts undergoing definitive RT. Methods: Eligibility included MIBC with plan to receive definitive RT +/- chemo. Fifteen patients were available for analysis (11 enrolled on trial from 2017-2018, plus an additional 4 received TraceIT off study). TraceIT was injected around the circumference of the tumor bed during pre-radiation maximal re-transurethral resection of bladder tumor (TURBT) for all pts, and again during the mid-radiation TURBT to improve visibility in n = 8 pts. The primary endpoint was assessment of interfraction motion on daily cone-beam CT (CBCTs) based on alignment to fiducial vs standard-of-care pelvic bone anatomy. Van Herk (VH) margin equation was used to determine the planning target volume margin optimized for the clinical target volume receiving at least 95%-prescription dose in 90% of pts. Recurrence rates and survival were estimated by Kaplan Meier. Toxicity was measured by CTCAE v4.03. Results: Patients underwent RT to a median total dose of 63.5 Gy (range 35.75-66.6), typically to an initial small pelvis field follow by a tumor boost. 14/15 received chemo with RT. Median TraceIT volume was 0.5cc (range 0.3-0.75) per injection site for 4 (range 4-6) sites per patient for total volume of 2cc (range 2-3). All pts demonstrated 100% visibility of TraceIT on initial simulation CT and day 1 CBCT. TraceIT visibility declined slightly over time, with 91.5% of patients having visible TraceIT by the end of the initial RT phase (usually week 5) and 82.5% by end of the boost phase. For the initial phase, alignment to fiducials over bone anatomy allowed for reduced VH margins (0.67cm vs 1.56cm). For the boost phase, the VH margin was similar between fiducial and bone alignment (1.01cm vs 0.96cm). There have been no acute or late complications from TraceIT placement. There were no grade grade (G) 4/5 toxicities; three pts had acute G 3 events, and three pts experienced late G 3 toxicity: two related to hematuria and one from ureteral stenosis. There were no late G≥2 GI toxicities. No patients have undergone cystectomy. At a median follow-up of 22 months, 2-yr OS was 79.1% and 2-yr PFS was 75.4%. Conclusions: TraceIT is safe and feasible for use in image-guided RT. TraceIT can increase the precision of RT by facilitating accurate target delineation of the bladder tumor bed and daily motion management which may allow for smaller radiation treatment margins to be used to reduce toxicity and to potentially facilitate safe dose escalation to tumor, which can improve tumor control. Clinical trial information: NCT03125226.

9040 Background: Patients with unresectable LA-NSCLC are treated with concurrent chemoradiation (CRT) and consolidation immunotherapy with survival that range from months to years or even decades. Early predictive biomarkers have potential to identify patients who are unlikely to benefit from continuing standard of care therapy and require a change in management. We investigated biomarkers that are widely available (PET/CT scan and plasma cytokine levels) to develop early predictors (mid-CRT) of survival in a phase II clinical trial of chemoradiation for LA-NSCLC. Methods: 37 Patients with AJCC v7 stage IIB-IIIB NSCLC were prospectively enrolled on the FLARE-RT trial (NCT02773238) from 2016-9. All patients underwent chemoradiation; 18 also received adjuvant durvalumab. PET/CT exams were performed at week 3 of CRT and response status was pre-defined by published metrics. 21 patients consented to peripheral blood collection at baseline and week 3, and plasma levels of 43 common inflammatory cytokines were measured. Bootstrapping over 100 iterations of the least absolute shrinkage and selection operator (LASSO) was performed to reduce feature dimensionality and guard against false discoveries. Cox regression of selected cytokine levels and PET response status, as well as time-dependent receiver-operating characteristic (ROC) analysis, were evaluated for associations to overall survival (OS). Results: Median follow-up was 18 months with 1-year OS 81% and PFS 52%. Mid-CRT PET response (as determined by pre-defined metrics) was strongly associated with OS (HR 5.6 [1.4-22.0], p = 0.015) after adjusting for radiation target volume, with 1-yr OS 94% for responders vs. 68% for non-responders (p = 0.017). Plasma TNFα level was also prognostic for OS (HR 1.9 [1.1-3.5], p = 0.030). TNFα retained significance for OS (HR 2.3 [1.2-4.6], p = 0.016) after adjusting for PET response. Bivariate mid-CRT PET response and TNFα generated a parsimonious model to predict OS (AUC = 0.85, 18-month horizon). Conclusions: Risk stratification for long-term survival after chemoradiation in patients with LA-NSCLC may be achievable based on mid-chemoradiation assessment of widely available biomarkers (PET imaging and plasma TNFα level). Combined functional imaging and peripheral blood biomarkers will be validated in a larger sample of our trial cohort, along with other independent patient populations. Clinical trial information: NCT02773238.

346 Background: Lymphopenia has been associated with survival and disease progression in esophageal cancer patients treated with chemoradiation (cRT). We previously published on our posterior-only proton therapy approach that maximally spares the heart and lungs, but at the cost of increased dose to the bone marrow in the vertebral bodies (VBs). We assessed hematologic toxicity in proton (PT) and IMRT treated patients and studied dosimetric parameters associated with hematologic toxicity. Methods: 35 patients treated with PT and 46 patients treated with IMRT for esophageal cancer between 2011-2018 were analyzed. Most patients were treated concurrently with carboplatin/paclitaxel to a median dose of 50.4 Gy. Lymphocyte, neutrophil and total leukocyte values while under treatment were recorded and graded per the CTCAE v4.03 toxicity scale, and the neutrophil-to-lymphocyte ratio (NLR) was computed. Mean dose and volumes (cc) receiving 5-50 Gy were calculated for the heart and VBs. A receiver-operator characteristic analysis was performed for univariate correlation between incidence of grade ≥3 hematotoxicity and dose-volume parameters. Results: Median follow-up was 36.1 months for all patients and the overall survival at 3 years was 57.5%. The rates of grade 3 or 4 hematologic toxicity in the PT group were 37.1% (leukopenia), 22.9% (neutropenia), and 80.0% (lymphopenia) versus 41.3%, 15.2% and 87.2%, respectively, for IMRT patients. There was a significant correlation between grade 4 lymphopenia and the heart V5, V10 and V20, but no significant correlation between VB doses with any hematotoxicity. Median NLR values and heart dose were higher in the IMRT group (9.17 vs 3.86 with PT, p = 0.0048; 10.5 vs 23.5Gy, p< 0.0001, respectively). There was a correlation between survival and NLR with a hazard ratio of 1.025 (CI 1.006 - 1.044). Conclusions: Low doses to the heart mediate severe lymphopenia in esophageal cancer patients treated with cRT. These data confirm the safety of the posterior-only proton approach without concern for increased hematologic toxicity despite higher vertebral body doses compared to IMRT. They also suggest that the blood pool is more important as a source of severe lymphopenia.

558 Background: High rates of local control are achievable with hypofractionated proton therapy with passive techniques for hepatocellular carcinoma (HCC), but may have limitations when tumors are adjacent to organs-at-risk (OARs), which may result in tumor underdosage and lead to inferior local control. We present the first reported series of HCC patients treated with pencil beam scanning (PBS) intensity-modulated proton therapy (IMPT) using a simultaneous-integrated boost and protection (SIB/SIP) technique to escalate tumor dose while protecting adjacent OARs. Methods: Twenty-five HCC patients were treated between 2015-2019 with a 15-fraction regimen using IMPT SIB/SIP. SIB/SIP dose levels generally ranged from 36.0-67.5 GyRBE to minimize dose to OARs at their respective dose-limiting thresholds (e.g. luminal gastrointestinal organs, chest wall). Radiation-induced liver disease (RILD) was defined by a Child-Pugh (CP) score increase of 2 or greater and/or any RTOG grade 3 enzyme elevation. Other toxicities were graded by CTCAEv5.0. Overall survival (OS), progression-free survival (PFS), and local control were calculated using the Kaplan-Meier method. Results: Patients most commonly had BCLC stage B or C disease (84%) and CP-A (80%) and ALBI grade 2 (60%) liver function. Median gross tumor volume (GTV) size and volume were 12.3 cm (range 2.17-20.57) and 461 cc (range 4.68-2439), and 32% had gross vascular invasion. Median mean and minimum dose delivered to the gross tumor volume (GTV) was 64.0 GyRBE (EQD2 76.1, BED 91.3, range 54.3-69.6) and 45.1 GyRBE (EQD2 48.9, BED 58.7, range 33.4-67.7), respectively. Median mean dose to liver minus GTV was 15.0 GyRBE (range, 8.2-19.6). 1-year OS, PFS, and local control were 66%, 32%, and 84%, respectively. No isolated local failures occurred. Two patients experienced RILD with no RILD-related deaths. Two grade 3 non-GI toxicities occurred: 1 rib fracture and 1 pneumonitis. No acute or late GI grade ≥2 occurred. Conclusions: In our series of HCC patients with large tumors near OARs, IMPT SIB/SIP allows for tumor dose escalation while sparing of OARs and results in favorable local control and acceptable toxicities.

444 Background: The albumin-bilirubin (ALBI) grade system was optimized to predict survival in hepatocellular carcinoma (HCC) patients; however, its applicability to patients treated with radiation therapy (RT) is unclear. We propose a novel analysis of the ALBI equation, designated ALBI-RT, to assess hepatotoxicity risk after RT. Methods: We retrospectively reviewed 48 consecutive HCC patients treated with RT (2013-2016). Raw ALBI values and Child-Pugh (CP) scores were calculated. Any patient deaths were examined for radiation induced liver disease (RILD). Raw ALBI was assessed as a continuous variable to perform ROC analyses to identify cutoffs for overall survival (OS) and RILD-specific survival (RILD-SS). Univariate predictors of OS and RILD-SS were evaluated using Cox regression to determine hazard ratios (HR). Results: Patient cohort was comprised of 60% CP-A and 39% CP-B/C. Median follow-up and OS was 13 and 10 months, respectively. There were 18 deaths with 6 ascribed to RILD. We identified a raw ALBI cutoff at -1.70 (AUC = 0.94, p = 0.008) that was predictive of RILD-SS with a sensitivity of 100% and specificity of 71%. Dichotimization of RILD-SS cutoff generated two ALBI-RT risk categories: low-risk A < -1.70 (n = 29) and high-risk B ≥ -1.70 (n = 18). ALBI-RT subdivided CP-A patients, identifying 14% at increased risk of RILD; conversely, ALBI-RT grade A identified 21% of CP-B/C patients at a decreased risk of RILD. For OS, raw ALBI as a continuous variable (HR 3.0, p = 0.02), and to a lesser degree ALBI-RT Grade B (HR 2.4, p = 0.06), performed similarly to traditional ALBI grade (HR 3.0, p = 0.01), and CP score (HR 1.4, p = 0.03). In contrast for RILD-SS, raw ALBI as a continuous variable (HR 25.1, p = 0.01) and ALBI-RT Grade B (HR 9.9, p = 0.04) were associated with an elevated relative risk of RILD than traditional ALBI grade (HR 5.8, p = 0.02) and CP score (HR 2.3, p = 0.003). Conclusions: ALBI-RT is a promising metric for pre-treatment assessment of HCC patients for predicting RILD-related death after RT with an elevated relative-risk compared to both CP and conventional ALBI grades. Future prospective evaluation and validation in independent data sets will strengthen the generalizability and utility of ALBI-RT.

361 Background: Radiation induced liver disease (RILD) is of critical concern in the treatment of hepatocellular carcinoma (HCC) with radiation therapy (RT). Variability exists in metrics used to define RILD with no consensus on which best predict for overall survival (OS) and RILD-specific survival (RILDSS). We examined the correlation between toxicity metrics and clinical outcomes in a heavily pre-treated population that received RT. Methods: The charts of 37 HCC patients treated from 2013 - 2015 were reviewed retrospectively. At baseline, 62% were Child-Pugh (CP)-A, 32% CP-B and 5% CP-C. The majority (59%) had prior liver-directed therapy (LDT), 43% received stereotactic body RT and 49% proton RT. Pre-treatment, toxicity ( ≤ 6 months from treatment) and outcomes data were collected. Deaths from RILD were scored. Pre-treatment factors and toxicity outcomes were assessed by univariate Cox models for association with OS and RILDSS. Statistically significant predictors formed the basis for stepwise multivariate Cox regression to retain independent predictors of survival. Results: At a median follow-up of 8 months, 14 patients had an increase in CP score ( ≥ 2, n = 7) and 3 had ≥ G3 RTOG transaminitis. There were 11 deaths, 5 from RILD. On univariate analysis (UA), tumor size, pre-treatment liver function, prior LDT and 5 toxicity metrics (CP score increases and transaminitis) were significantly associated with OS. An increase of ≥ 1 CP score (HR 22.7, p = 0.005), pre-treatment ALBI grade (HR 6.0, p = 0.02) and tumor size (HR 1.2, p = 0.01) were independent predictors of OS on multivariate analysis (MVA). Similar factors were associated with RILDSS on UA, including ≥ 2 CP score increase and ≥ G3 ALT elevation; however, only pre-treatment CP score (HR 4.0, p = 0.01) and tumor size (HR 1.5, p = 0.03) were independently predictive on MVA. Conclusions: Pre-treatment liver functional status and tumor size were highly predictive of OS and RILDSS, suggesting that baseline functional hepatic reserve is the primary determinant in developing fatal RILD rather than post-RT changes in liver function. Further work is needed to define dosimetric parameters and pre-treatment factors that predict RILD toxicity.