Patients diagnosed with metastatic sarcoma have limited options for achieving both local and distant tumor control. While SBRT can achieve local control, distant response rates remain low. There is limited evidence demonstrating the safety and efficacy for combining SBRT with concurrent PD-1 checkpoint blockade in metastatic sarcoma. In this prospective case-series, we examined five patients with metastatic sarcoma on pembrolizumab treated concurrently with SBRT from July 1, 2016-October 30, 2018. Acute and chronic toxicity were recorded using Common Terminology Criteria for Adverse Events (CTCAE, version 5.0). SBRT-treated tumor control was assessed using Response Evaluation Criteria in Solid Tumors (RECIST version 1.1). With median follow-up of 14.9 months, three patients with undifferentiated pleomorphic sarcoma, one with intimal, and one with chondroblastic osteosarcoma received SBRT with concurrent pembrolizumab to 10 sites of metastatic disease. No grade 5 toxicities were observed. There was a single incidence of transient grade 4 lymphopenia which resolved without intervention. Grade 3 toxicities included anemia, thrombocytopenia, lymphopenia and colitis. One tumor demonstrated local progression after SBRT, and all others remained stable or with response. In conclusion, combining SBRT with PD-1 inhibition appeared to be safe in this patient population. Expected high rates of treated-tumor local control after SBRT were observed. Two of five patients demonstrated either enhanced local tumor regression, or possible abscopal effect.

Purpose: This study aimed to investigate the feasibility of stereotactic body radiation therapy (SBRT) as salvage therapy for locally recurrent esophageal cancer. We hypothesized that SBRT would provide durable treated tumor control with minimal associated toxicity in patients with progressive disease after definitive radiation, chemotherapy, and surgical resection. Methods: This single-institution retrospective study assessed outcomes in patients who received SBRT for locoregional failure of esophageal cancer after initial curative-intent treatment. Only patients who had received neoadjuvant chemoradiation (≥41.4 Gy) for esophageal cancer were selected. Subsequent surgical resection was optional but institutional follow-up by an oncologist was required. The primary endpoints of this study were gastrointestinal and constitutional toxicity, scored with the Common Terminology Criteria for Adverse Events v5.0. A secondary outcome, treated-tumor control, was assessed with RECIST v1.1. Results: Nine patients (11 locoregional recurrences) treated with SBRT were reviewed, with a median follow-up time of 10.5 months. Most patients initially presented with T3 (88.9%), N1 (55.6%), moderately differentiated (66.7%) adenocarcinoma (88.9%), and had received a median 50.4 Gy delivered over 28 fractions with concurrent carboplatin/paclitaxel chemotherapy followed by surgical resection. Median time to recurrence was 16.3 months. Median total dose delivered by SBRT was 27.5 Gy (delivered in five fractions). Two patients experienced acute grade 1 fatigue and vomiting. No patient experienced grade 3 or higher toxicity. One patient experienced failure in the SBRT treatment field at 5.8 months after treatment and six patients developed distant failure. The median progression-free survival time for SBRT-treated tumors was 5.0 months, and median overall survival time was 12.9 months. Conclusions: This single-institution study demonstrated the feasibility of SBRT for locoregional recurrence of esophageal cancer with minimal treatment-related toxicity and high rates of treated tumor control. Prospective studies identifying ideal salvage SBRT candidates for locoregional failure as well as validating its safety are needed.

PURPOSE: The majority of colorectal cancers are resistant to cancer immune checkpoint inhibitors. Ionizing radiation (IR) and several radiosensitizers, including PARP inhibitors, can enhance responsiveness to immune checkpoint inhibitors by potentially complementary mechanisms of action. We assessed the ability of radiation and PARP inhibition to induce proimmunogenic changes in tumor cells and enhance their in vivo responsiveness to anti-PD-1 antibodies. METHODS AND MATERIALS: We performed a candidate drug screen and used flow cytometry to assess effects of the PARP inhibitor veliparib on IR-mediated changes in MHC-1 antigen presentation and surface localization of immune-modulating proteins including PD-L1 and calreticulin in colorectal cancer tumor models. Reverse transcription polymerase chain reaction was used to assess the effects of veliparib and radiation on the expression of proinflammatory and immunosuppressive cytokines. The ability of concurrent PARP inhibition and subablative doses of radiation therapy to enhance in vivo responsiveness to anti-PD-1 antibodies was assessed using unilateral flank-tumor models with or without T-cell depletion. RESULTS: Veliparib was a potent radiosensitizer in both cell lines. Radiation increased surface localization of MHC-1 and PD-L1 in a dose-dependent manner, and veliparib pretreatment significantly enhanced these effects with high (8 Gy) but not with lower radiation doses. Enhancement of MHC-1 and PD-L1 surface localization by IR and IR+ veliparib remained significant 1, 3, and 7 days after treatment. IR significantly increased delayed tumoral expression of proinflammatory cytokines interferon-Ƴ and CXCL10 but had no significant effect on the expression of IL-6 or TGF-β. Concurrent administration of veliparib and subablative radiation therapy (8 Gy × 2) significantly prolonged anti-PD-1-mediated in vivo tumor growth delay and survival in both tumor models. Moreover, these effects were more pronounced in the microsatellite instability-mutated MC38 tumor model. Enhancement of anti-PD-1 mediated tumor growth delay with veliparib and IR was attenuated by CD8+ T-cell depletion. CONCLUSIONS: We provide preclinical evidence for a novel therapeutic strategy to enhance responsiveness of colorectal tumors to immune checkpoint inhibitors.

Non-small-cell lung cancer (NSCLC) represents approximately 80-85% of lung cancer diagnoses and is the leading cause of cancer-related death worldwide. Recent studies indicate that image-based radiomics features from positron emission tomography/computed tomography (PET/CT) images have predictive power for NSCLC outcomes. To this end, easily calculated functional features such as the maximum and the mean of standard uptake value (SUV) and total lesion glycolysis (TLG) are most commonly used for NSCLC prognostication, but their prognostic value remains controversial. Meanwhile, convolutional neural networks (CNN) are rapidly emerging as a new method for cancer image analysis, with significantly enhanced predictive power compared to hand-crafted radiomics features. Here we show that CNNs trained to perform the tumor segmentation task, with no other information than physician contours, identify a rich set of survival-related image features with remarkable prognostic value. In a retrospective study on pre-treatment PET-CT images of 96 NSCLC patients before stereotactic-body radiotherapy (SBRT), we found that the CNN segmentation algorithm (U-Net) trained for tumor segmentation in PET and CT images, contained features having strong correlation with 2- and 5-year overall and disease-specific survivals. The U-Net algorithm has not seen any other clinical information (e.g. survival, age, smoking history, etc.) than the images and the corresponding tumor contours provided by physicians. In addition, we observed the same trend by validating the U-Net features against an extramural data set provided by Stanford Cancer Institute. Furthermore, through visualization of the U-Net, we also found convincing evidence that the regions of metastasis and recurrence appear to match with the regions where the U-Net features identified patterns that predicted higher likelihoods of death. We anticipate our findings will be a starting point for more sophisticated non-intrusive patient specific cancer prognosis determination. For example, the deep learned PET/CT features can not only predict survival but also visualize high-risk regions within or adjacent to the primary tumor and hence potentially impact therapeutic outcomes by optimal selection of therapeutic strategy or first-line therapy adjustment.

PURPOSE: For inoperable stage I (T1-T2N0) small cell lung cancer (SCLC), national guidelines recommend chemotherapy with or without conventionally fractionated radiation therapy. The present multi-institutional cohort study investigated the role of stereotactic ablative radiation therapy (SABR) for this population. METHODS AND MATERIALS: The clinical and treatment characteristics, toxicities, outcomes, and patterns of failure were assessed in patients with histologically confirmed stage T1-T2N0M0 SCLC. Kaplan-Meier analysis was used to evaluate the survival outcomes. Univariate and multivariate analyses identified predictors of outcomes. RESULTS: From 24 institutions, 76 lesions were treated in 74 patients (median follow-up 18 months). The median age and tumor size was 72 years and 2.5 cm, respectively. Chemotherapy and prophylactic cranial irradiation were delivered in 56% and 23% of cases, respectively. The median SABR dose and fractionation was 50 Gy and 5 fractions. The 1- and 3-year local control rate was 97.4% and 96.1%, respectively. The median disease-free survival (DFS) duration was 49.7 months. The DFS rate was 58.3% and 53.2% at 1 and 3 years, respectively. The median, 1-year, and 3-year disease-specific survival was 52.3 months, 84.5%, and 64.4%, respectively. The median, 1-year, and 3-year overall survival (OS) was 17.8 months, 69.9%, and 34.0% respectively. Patients receiving chemotherapy experienced an increased median DFS (61.3 vs 9.0 months; P=.02) and OS (31.4 vs 14.3 months; P=.02). The receipt of chemotherapy independently predicted better outcomes for DFS/OS on multivariate analysis (P=.01). Toxicities were uncommon; 5.2% experienced grade ≥2 pneumonitis. Post-treatment failure was most commonly distant (45.8% of recurrence), followed by nodal (25.0%) and elsewhere lung (20.8%). The median time to each was 5 to 7 months. CONCLUSIONS: From the findings of the largest report of SABR for stage T1-T2N0 SCLC to date, SABR (≥50 Gy) with chemotherapy should be considered a standard option.

PURPOSE: Standard treatment for glioblastoma (GBM) includes surgery, radiation therapy (RT), and temozolomide (TMZ), yielding a median overall survival (OS) of approximately 14 months. Preclinical models suggest that pharmacologic ascorbate (P-AscH-) enhances RT/TMZ antitumor effect in GBM. We evaluated the safety of adding P-AscH- to standard RT/TMZ therapy. PATIENTS AND METHODS: This first-in-human trial was divided into an RT phase (concurrent RT/TMZ/P-AscH-) and an adjuvant (ADJ) phase (post RT/TMZ/P-AscH- phase). Eight P-AscH- dose cohorts were evaluated in the RT phase until targeted plasma ascorbate levels were achieved (≥20 mmol/L). In the ADJ phase, P-AscH- doses were escalated in each subject at each cycle until plasma concentrations were ≥20 mmol/L. P-AscH- was infused 3 times weekly during the RT phase and 2 times weekly during the ADJ phase continuing for six cycles or until disease progression. Adverse events were quantified by CTCAE (v4.03). RESULTS: Eleven subjects were evaluable. No dose-limiting toxicities occurred. Observed toxicities were consistent with historical controls. Adverse events related to study drug were dry mouth and chills. Targeted ascorbate plasma levels of 20 mmol/L were achieved in the 87.5 g cohort; diminishing returns were realized in higher dose cohorts. Median progression-free survival (PFS) was 9.4 months and median OS was 18 months. In subjects with undetectable MGMT promoter methylation (n = 8), median PFS was 10 months and median OS was 23 months. CONCLUSIONS: P-AscH-/RT/TMZ is safe with promising clinical outcomes warranting further investigation.

Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing.