Search

Merkel cell carcinoma (MCC) is a rare but clinically aggressive cancer with a high mortality rate. In recent years, antibodies blocking the interactions among PD-1 and its ligands have generated durable tumor regressions in patients with advanced MCC. However, there is a paucity of data regarding effective therapy for patients whose disease is refractory to PD-1 pathway blockade. This retrospective case series describes a heterogeneous group of patients treated with additional immune checkpoint blocking therapy after MCC progression through anti-PD-1. Among 13 patients treated with anti-CTLA-4, alone or in combination with anti-PD-1, objective responses were seen in 4 (31%). Additionally, one patient with MCC refractory to anti-PD-1 and anti-CTLA-4 experienced tumor regression with anti-PD-L1. Our report – the largest case series to date describing this patient population – provides evidence that sequentially-administered salvage immune checkpoint blocking therapy can potentially activate anti-tumor immunity in patients with advanced anti-PD-1-refractory MCC and provides a strong rationale for formally testing these agents in multicenter clinical trials. Additionally, to the best of our knowledge, our report is the first to demonstrate possible anti-tumor activity of second-line treatment with a PD-L1 antibody in a patient with anti-PD-1-refractory disease. Electronic supplementary material The online version of this article (10.1186/s40425-019-0661-6) contains supplementary material, which is available to authorized users.

Colitis is a frequent, clinically-significant immune-related adverse event caused by anti-programmed death-1 (PD-1). The clinical features, timing, and management of colitis with anti-PD-1-based regimens are not well-characterized. Patients with advanced melanoma that received either anti-PD-1 monotherapy (“monotherapy”) or combined with ipilimumab (“combination therapy”) were screened from 8 academic medical centers, to identify those with clinically-relevant colitis (colitis requiring systemic steroids). Of 1261 patients who received anti-PD-1-based therapy, 109 experienced colitis. The incidence was 3.2% (30/937) and 24.4% (79/324) in the monotherapy and combination therapy cohorts, respectively. Patients with colitis from combination therapy had significantly earlier symptom onset (7.2 weeks vs 25.4 weeks, p

Background ONC201 is a small molecule antagonist of DRD2, a G protein-coupled receptor overexpressed in several malignancies, that has prolonged antitumor efficacy and immunomodulatory properties in preclinical models. The first-in-human trial of ONC201 previously established a recommended phase II dose (RP2D) of 625 mg once every three weeks. Here, we report the results of a phase I study that evaluated the safety, pharmacokinetics (PK), and pharmacodynamics (PD) of weekly ONC201. Methods Patients ≥ 18 years old with an advanced solid tumor refractory to standard treatment were enrolled. Dose escalation proceeded with a 3 + 3 design from 375 mg to 625 mg of ONC201. One cycle, also the dose-limiting toxicity (DLT) window, was 21 days. The primary endpoint was to determine the RP2D of weekly ONC201, which was confirmed in an 11-patient dose expansion cohort. Results Twenty patients were enrolled: three at 375 mg and 17 at 625 mg of ONC201. The RP2D was defined as 625 mg with no DLT, treatment discontinuation, or dose modifications due to drug-related toxicity. PK profiles were consistent with every-three-week dosing and similar between the first and fourth dose. Serum prolactin and caspase-cleaved cytokeratin-18 induction were detected, along with intratumoral integrated stress response activation and infiltration of granzyme B+ Natural Killer cells. Induction of immune cytokines and effectors was higher in patients who received ONC201 once weekly versus once every three weeks. Stable disease of > 6 months was observed in several prostate and endometrial cancer patients. Conclusions Weekly, oral ONC201 is well-tolerated and results in enhanced immunostimulatory activity that warrants further investigation. Trial registration NCT02250781 (Oral ONC201 in Treating Patients With Advanced Solid Tumors), NCT02324621 (Continuation of Oral ONC201 in Treating Patients With Advanced Solid Tumors). Electronic supplementary material The online version of this article (10.1186/s40425-019-0599-8) contains supplementary material, which is available to authorized users.

e21551 Background: Combined BRAF/MEK inhibition results in improved progression free and overall survival in BRAF mutated melanoma, but significant response is not universally observed. TERT promoter activating mutations often co-occur with BRAF mutations and have been associated with aggressive features and poorer prognosis. The TERT promotor inhibits apoptosis via a mechanism dependent upon BRAF mutant MAPK activation. Preclinical data in mouse models suggests that BRAF/TERT genetic duet melanomas are associated with improved response to BRAF/MEK inhibition as compared with BRAF mutant/TERT-WT melanomas. Methods: We performed a single center retrospective analysis of adults with melanoma with confirmed BRAF mutations +/- TERT promoter mutations. Responses and progression free survival in response to BRAF/MEK inhibition was assessed. Differences in RR and PFS were compared using Kaplan-Meier and Log-Rank. Results: 52 cases of BRAF/TERT genetic duet and BRAF mutated/TERT-WT melanomas were assessed. A total of 24 patients received BRAF/MEK inhibitors over the course of treatment meeting criteria for study inclusion; 9 (37.5%) BRAF/TERT genetic duet and 15 (62.5%) BRAF mutated/TERT-WT. BRAF V600E was present in 19/24 (79.2%) and V600K in 5/24 (20.8%). In the genetic duets, TERT -146C > T was present in 4/9 (44.4%), -124C > T in 2/9 (22.2%), -139_-138CC > TT in 2/9 (22.2%), and a SNV in 1/9 (11.1%). Mean age at diagnosis was 56 ± 13.5 years and 62.5% were male. ECOG PFS was 0-1 in 15/24 (62.5%), 2-3 in 6/24 (25%), and unreported in 3/24 (12.5%). Mean LDH at start of therapy was 391 (range 81-1664). At initial diagnosis 20.8% were Stage I, 25% Stage II, 37.5% Stage III, and 16.7% Stage IV. Two or more sites of disease were present in 10/24 (41.7%) and 2/24 (8.3%) had CNS metastases. BRAF/MEK directed therapy was first line in 6/24 (25%) of patients, others received prior immunotherapy. No significant differences between groups were observed in baseline demographics, disease state at diagnosis, or treatment history. In BRAF/TERT genetic duet melanomas CR was observed in 1/9 (11.1%), PR in 7/9 (77.8%), and NR in 1/9 (11.1%). In BRAF mutated/TERT-WT CR was observed in 3/15 (20%), PR in 11/15 (73.3%), and NR in 1/15 (6.7%). BRAF/TERT genetic duets were observed to initially have somewhat better PFS on first exposure to BRAF/MEK directed therapy but the PFS curves crossed at about 5 months with no significant difference observed overall (p = 0.40). Conclusions: This study is the first to report on outcomes of BRAF/MEK directed therapy in BRAF/TERT genetic duet vs BRAF mutated/TERT-WT melanomas in humans. While preclinical data from mouse models observed an improved response to BRAF/MEK inhibition in genetic duet tumors, no significant difference was observed. Our study is limited by small sample size. A multicenter analysis may be of interest to better understand the effects of BRAF inhibition in patients with BRAF/TERT genetic duet melanoma.

e22034 Background: Immune checkpoint inhibitors (ICIs) improve survival outcomes in metastatic melanoma and non-small cell lung cancer (NSCLC). Preclinical evidence suggests that overexpression of cyclooxygenase-2 (COX2) in tumors facilitates immune evasion through prostaglandin E2 production and that COX inhibition synergizes with ICIs to promote antitumor T-cell activation. This study investigates whether concurrent COX inhibitor (COXi) use during ICI treatment is associated with improved time-to-progression (TTP), objective response rate (ORR) and overall survival (OS) in metastatic melanoma and NSCLC patients. Methods: We retrospectively reviewed 90 metastatic melanoma and 37 metastatic NSCLC patients, treated only with ICI between 2011 and 2019. Differences in TTP and OS by COXi use were compared using Kaplan-Meier and Cox regression. Interaction between COXi use and neutrophil-lymphocyte ratio (NLR) was examined. Independent radiology review per RECIST v1.1 was performed. Results: For melanoma patients, median TTP was significantly prolonged in COXi users (245 v 100.5 days, p = .0002). On multivariate analysis (MVA), COXi use associated with increased TTP (HR 0.36, 95% CI: 0.2 – 0.66, p = 0.001) when adjusted for age, pretreatment NLR, and gender. For NSCLC patients, COXi use also associated with increased TTP on MVA (HR 0.45; 95% CI: 0.21 – 0.97; p = 0.042) adjusted for age. ORR at 6-months was significantly higher in patients who received concurrent COXi compared to those that did not in both melanoma (58.6% v 19.2%, p = 0.0005) and NSCLC (73.7% v 33.3%, p = 0.036) cohorts. COXi use was associated with improved OS in NSCLC patients on MVA (HR 0.3; 95% CI: 0.12 – 0.78; p = 0.013), but not in melanoma patients. In the melanoma cohort, high pretreatment NLR of > 5 associated with decreased TTP on MVA (HR 3.21, 95% CI: 1.64 – 6.3; p = 0.0007). However, for melanoma patients with high NLR > 5, COXi use significantly associated with increased TTP (HR 0.08, 95% CI: 0.03 – 0.25); but in melanoma patients with low NLR ≤5, there was no effect on TTP (HR 0.65, 95% CI: 0.32-1.32). Similar outcomes were found in an adjusted melanoma cohort after RECIST review. Conclusions: Our study suggests that use of COXi concurrent with ICI is significantly associated with longer TTP and improved ORR at 6-months in metastatic melanoma and NSCLC patients. NSCLC patients using COXi had prolonged OS. Furthermore, COXi use appears to reverse the negative prognostic effect of a high NLR by prolonging TTP in melanoma patients.

79 Background: The Metabotropic Glutamate Receptor 1 (GRM1) is overexpressed in many solid tumors. It activates MAPK and PI3K/AKT pathways and promotes an immune-suppressive tumor microenvironment. In an immunocompetent melanoma mouse model, GRM1 blockade using riluzole or its pro-drug troriluzole (BHV-4157) and PD-1 blockade inhibit tumor growth in an additive manner. We conducted a phase Ib trial of troriluzole and the PD-1 antibody nivolumab. Methods: Patients with advanced or refractory solid tumors were treated with increasing doses of troriluzole. Troriluzole monotherapy was given for a 14-day lead-in period and then patients received troriluzole in combination with nivolumab 240 mg IV every 2 weeks. Results: We enrolled 14 patients with advanced solid tumors (melanoma=3, NSCLC=3, renal cell cancer=2, NSCLC=2, head and neck cancer=2). Eleven patients had prior therapy with anti-PD-(L)1. PK sampling demonstrated that the prodrug was cleaved efficiently without regard to food (Table). The most common TEAEs (all grades) occurring in >40% of patients were transaminitis, increased lipase and nausea. DLT occurred in 3 patients: 1) grade 3 anorexia, 2) grade 3 fatigue and, 3) atrial fibrillation. The MTD was determined to be troriluzole 140 mg QAM + 280 mg PO QHS. The response rate was 7 % (1/14); this occurred in a PD-L1 treated patient. The 6-month PFS rate was 21%. Correlative studies will be presented. Conclusions: The combination of troriluzole and nivolumab was safe and well-tolerated in this highly PD-1 refractory population. Clinical trial information: NCT03229278. [Table: see text]

This cross-sectional study weighs the proportional associations of rising drug costs and use with increased Medicare Part D spending for a cohort of oral anticancer drugs used from 2013 through 2017.

Background Metastatic melanoma (mM) and renal cell carcinoma (mRCC) are often treated with anti-PD-1 based therapy, however not all patients respond and further therapies are needed. High dose interleukin-2 (HD IL-2) can lead to durable responses in a subset of mM and mRCC patients. The efficacy and toxicity of HD IL-2 therapy following anti-PD-1 or anti-PD-L1 therapy have not yet been explored. Methods Reports on mM and mRCC patients who had received HD IL-2 after PD-1 or PD-L1 inhibition were queried from the PROCLAIMSM database. Patient characteristics, toxicity and efficacy were analyzed. Results A total of 57 patients (40 mM, 17 mRCC) were treated with high dose IL-2 after PD-1 or PD-L1 inhibition and had data recorded in the PROCLAIM database. The best overall response rate to HD IL-2 was 22.5% for mM (4 complete response (CR), 5 partial responses (PRs)) and 24% for mRCC (2 CRs, 2 PRs). The toxicity related to HD IL-2 observed in these patients was similar to that observed in patients treated with HD IL-2 without prior checkpoint blockade. One patient who had received prior PD-L1 blockade developed drug induced pneumonitis with HD IL-2 requiring steroid therapy. Conclusion In this retrospective analysis, HD IL-2 therapy displayed durable antitumor activity in mM and mRCC patients who progressed following treatment with PD-1 and PD-L1 inhibition. The toxicities were generally manageable and consistent with expectations from HD IL-2 but physicians should watch for immune related toxicities such as pneumonitis. This analysis supports the development of randomized prospective trials to assess the proper sequencing and combination of immune checkpoint blockade and cytokine therapy. Electronic supplementary material The online version of this article (10.1186/s40425-019-0522-3) contains supplementary material, which is available to authorized users.

Interferon (IFN)-λ, a type III interferon (IFN), is a member of a new family of pleotropic cytokines that share high similarity with classical IFNs α and β (IFN-α/β), type I IFNs. IFN-λ acts as an antiviral agent and displays distinct biological functions, including tumor suppression. Although it activates the common Janus kinase (JAK) and signal transducer and activator of transcription (STAT) pathways, similar to IFN-α/β, IFN-λ differentially induces the expression of IFN-stimulated genes (ISGs). Novel evidence indicates that IFN-λ acts quite differently from IFN-α/β under both homeostasis and pathological situations. In contrast to IFN-α/β, IFN-λ is not involved in over-stimulation of the immune response or exacerbation of inflammation. However, the emergence of unexpected characteristics of IFN-λ, in the control of inflammation and promotion of immune suppression and cancer, reveals novel challenges and offers more strategic opportunities in the context of cancer and beyond. In this article, we discuss new evidence and potential consequences associated with the biology of IFN-λ and provide a different vision for building novel therapeutic strategies in oncology.

86 Background: Pain is a common symptom among patients with cancer and impacts performance status and quality of life. The OCM was implemented at Rutgers Cancer Institute of New Jersey (RCINJ) in July 2016 with the goal of improving many quality metrics in cancer care including documentation of pain and a plan of care for pain. Hypothesis: Documentation of pain and a plan of care for pain improved after implementation of the OCM at RCINJ. Improved documentation would correlate with a decrease in reported quantitative pain scores. Methods: Data from patients enrolled in OCM during its first year of implementation (post OCM) was compared to data from a control cohort the preceding year (pre OCM). The initial progress note and EMR flowsheet in each year were reviewed to determine pain documentation. For patients reporting a pain score ≥ 1 on a 0-10 scale, the association between documentation of a plan of care and improvement in quantitative pain score on a subsequent visit was analyzed using Chi square testing. Results: A total of 260 patient charts were analyzed. A quantitative pain score was documented in 99% of patients in the pre OCM group and 100% in the post OCM group. For those with a pain score ≥ 1, documentation of a plan of care increased from 43% to 55% after OCM implementation. Patients whose charts contained documented plans of care for pain were less likely to have decreased pain scores at a subsequent visit (51% vs 76%). Conclusions: (1) Documentation of a quantitative pain score was completed in nearly all patients before and after the implementation of OCM. (2) There was a non-statistically significant increase in documentation of plan of care for pain after OCM implementation. (3) Documentation of a plan of care was not associated with a decreased pain score at a subsequent visit. This may be the result of small sample size or related to progressive cancer-related pain or inadequate pain intervention. [Table: see text]

e21029 Background: In tumors, cyclooxygenase (COX) activity leads to biosynthesis of prostaglandin E2 which can inhibit T-cell and dendritic cell function, thereby allowing immune evasion. Preclinical models have shown therapeutic synergy between COX inhibitors (COXi) and anti-PD-1 inhibition (PD-1i) immunotherapy (IT). We hypothesized that patients (pts) treated with checkpoint blockade IT and concurrent COXi would have improved responses compared to those treated with IT and no COXi. Methods: We retrospectively analyzed 65 metastatic melanoma pts who received IT (PD-1i/PD-L1i/CTLA4i) between 5/2011-11/2017 on their 1st course of IT. Of these pts, 38.5% (N = 25) took concurrent COXi (aspirin, NSAIDs, or selective COX-2i) and 61.5% (N = 40) did not. We compared differences in baseline characteristics and time to progression (TTP) between IT-treated melanoma pts who did or did not receive COXi, using the Kaplan-Meier method and Cox regression. TTP was defined as time from IT start to progression of disease (POD) as determined by the treating physician based on CT or PET/CT scans, or last known alive. Results: Mean age was 66 yrs; male: 64.6%; BRAF mutant: 29.2%, ECOG 0/1/2: 60%/35.4%/4.6%; steroid use: 47.7% during IT; and prior non-checkpoint blockade IT: 13.8%. IT was stopped due to POD in 53.8% of pts, IT toxicity 6.2%, completed therapy 24.6%, and treatment holiday 15.4%. There were no significant differences in baseline characteristics between pts who did or did not receive COXi during IT, except for mean age (70 (COXi) versus 63 yrs (no COXi), p = 0.019). Pts that received COXi had longer TTP (median 245 days) vs without COXi (median 85 days, p = 0.0002). On multivariate analysis, COXi was associated with longer TTP (HR 0.34, p = 0.0004) after accounting for age, gender, steroid use, and ECOG PS ≥1. Immune-related toxicity was not significantly different with or without COXi (p = 0.73). Conclusions: Our findings suggest that in metastatic melanoma pts treated with IT, concurrent COXi use may result in a longer TTP, without increasing toxicity. These findings warrant further study to continue to augment outcomes for pts.

Ipilimumab, an antibody that enhances T-cell activation, may augment immunogenicity of tumor cells that are injured by radiation therapy. We hypothesized that patients with melanoma brain metastasis treated with both ipilimumab and radiotherapy would have improved overall survival, and that the sequence of treatments may affect disease control in the brain. We analyzed the clinical and radiographic records of melanoma patients with brain metastases who were treated with whole brain radiation therapy or stereotactic radiosurgery between 2005 and 2012. The hazard ratios for survival were estimated to assess outcomes as a function of ipilimumab use and radiation type. Seventy patients were identified, 33 of whom received ipilimumab and 37 who did not. The patients who received ipilimumab had a censored median survival of 18.3 months (95% confidence interval 8.1–25.5), compared with 5.3 months (95% confidence interval 4.0–7.6) for patients who did not receive ipilimumab. Ipilimumab and stereotactic radiosurgery were each significant predictors of improved overall survival (hazard ratio = 0.43 and 0.45, with P = 0.005 and 0.008, respectively). Four of 10 evaluable patients (40.0%) who received ipilimumab prior to radiotherapy demonstrated a partial response to radiotherapy, compared with two of 22 evaluable patients (9.1%) who did not receive ipilimumab. Ipilimumab is associated with a significantly reduced risk of death in patients with melanoma brain metastases who underwent radiotherapy, and this finding supports the need for multimodality therapy to optimize patient outcomes. Prospective studies are needed and are underway.

O1 IL-15 primes an mTOR-regulated gene-expression program to prolong anti-tumor capacity of human natural killer cells #### Andreas Lundqvist1, Vincent van Hoef1, Xiaonan Zhang1, Erik Wennerberg2, Julie Lorent1, Kristina Witt1, Laia Masvidal Sanz1, Shuo Liang1, Shannon Murray3, Ola Larsson1

Purpose: ONC201 is a small-molecule selective antagonist of the G protein–coupled receptor DRD2 that is the founding member of the imipridone class of compounds. A first-in-human phase I study of ONC201 was conducted to determine its recommended phase II dose (RP2D). Experimental Design: This open-label study treated 10 patients during dose escalation with histologically confirmed advanced solid tumors. Patients received ONC201 orally once every 3 weeks, defined as one cycle, at doses from 125 to 625 mg using an accelerated titration design. An additional 18 patients were treated at the RP2D in an expansion phase to collect additional safety, pharmacokinetic, and pharmacodynamic information. Results: No grade >1 drug-related adverse events occurred, and the RP2D was defined as 625 mg. Pharmacokinetic analysis revealed a Cmax of 1.5 to 7.5 μg/mL (∼3.9–19.4 μmol/L), mean half-life of 11.3 hours, and mean AUC of 37.7 h·μg/L. Pharmacodynamic assays demonstrated induction of caspase-cleaved keratin 18 and prolactin as serum biomarkers of apoptosis and DRD2 antagonism, respectively. No objective responses by RECIST were achieved; however, radiographic regression of several individual metastatic lesions was observed along with prolonged stable disease (>9 cycles) in prostate and endometrial cancer patients. Conclusions: ONC201 is a selective DRD2 antagonist that is well tolerated, achieves micromolar plasma concentrations, and is biologically active in advanced cancer patients when orally administered at 625 mg every 3 weeks. Clin Cancer Res; 23(15); 4163–9. ©2017 AACR.

Purpose: Metastatic breast cancer patients continue to have poor outcomes on current treatments, and thus, novel therapies in combination with standard treatments are needed. Towards this we studied the pre-clinical effects of combining ONC201, a first-in-class imipridone (which works via activation of ER stress, upregulation of TRAIL, and activation of death receptor 5 [DR5]) with radiation therapy (RT). Since the DR5 pathway is also required for RT-induced apoptosis, we hypothesized that addition of ONC201 could increase the effectiveness of RT. Pertinent Experimental Procedures: Human and mouse breast cancer cell lines were used to study in vitro cancer cell death (by AlamarBlue and clonogenic assays). Patient-derived breast cancer tissue and mouse cell lines were used to study in vivo tumor growth kinetics. Cell lines and mice were administered radiation (0-8 Gy) delivered via a Gammacell 40 exactor. ONC201 was given at does 0-10 µM. Mechanisms of tumor cell death were elucidated using western blots of cleaved PARP (cPARP) and caspase 3 (cC3). Immune response mechanisms were elucidated using in vivo cell-specific depletion. Summary of Data: Multiple breast cancer lines (MB231, MB468, and 4T1) exhibited decreased cell viability after combination therapy compared to either single therapy. In particular, ON201 (1 µM) followed by RT (8 Gy) resulted in a decreased MB468 breast cancer cell viability compared to no treatment (36.9%, P Citation Format: Sachin R. Jhawar, Devora Schiff, Hao Wu, Aditya Thandoni, Suemair Hassan, Joshua Allen, Martin Stogniew, Rohinton Tarapore, Wolfgang Oster, Mark Stein, Ann W. Silk, Sharad Goyal, Bruce G. Haffty, Andrew Zloza. Combination ONC201 and radiation therapy in the treatment of breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2765.

TPS219 Background: Talimogene laherparepvec, a modified herpes virus agent, induces a response in 65% of injected melanoma tumors. The combination of talimogene laherparepvec with ipilimumab or pembrolizumab appears promising in clinical trials of advanced melanoma. Talimogene laherparepvec-based therapy may be effective in other cancers of the skin and lymph nodes that are anatomically accessible for intratumoral injection. Methods: This phase II study will evaluate intratumoral talimogene laherparepvec monotherapy in 4 parallel disease cohorts: 1) Refractory T cell and NK cell lymphomas including cutaneous T cell lymphoma, 2) Merkel cell carcinoma 3) Cutaneous squamous cell carcinoma and 4) Other advanced/refractory non-melanoma skin cancers. Lymphoma patients must be refractory to or intolerant of all standard life-prolonging therapies. Skin cancer patients must be advanced/unresectable or refractory to one or more treatments including surgery, radiation therapy, or medical therapy. Prior PD-1-directed therapy is allowed. If an objective response is not achieved by Week 12, the PD-1 blocking antibody nivolumab will be added. The primary endpoint is the response rate with talimogene laherparepvec and secondary endpoints include response rate with the combination and overall survival. Using a two-stage design, if 1 or more response is observed in the first 9 patients in each parallel cohort, 8 additional patients will be accrued for a total sample size of 36 to 68 patients across the 4 disease cohorts. Tumor biopsies of injected lesions are mandatory at baseline and Week 6, and optional at Week 16 and the time of progression. Optional biopsies of non-injected lesions (when applicable) at Week 6 and 16 will be analyzed to identify biomarkers of systemic immunity. Tumor tissue and/or blood will be assayed for PD-L1 expression, RNA profiling, immune cell profiling, HVEM, NECTIN 1/2, IDO, tryptophan and L-kynurenine, mutational load, TIL TCR clonality, and prior exposure to herpes simplex type 1 virus and Merkel cell polyomavirus. Clinical trial information: NCT02978625.

TPS80 Background: The Metabotropic Glutamate Receptor 1 (GRM1) is expressed in 60-100% of human melanomas, breast cancers, and other solid tumors. Riluzole, an FDA-approved drug for ALS, inhibits GRM1 signal transduction. A phase 0 trial in melanoma patients demonstrated that riluzole suppressed signaling through the MAPK and PI3K/AKT pathways but no objective responses were seen in a phase 2 study.The clinical use of riluzole outside of ALS has been limited by: (i) poor oral bioavailability, (ii) extensive first-pass hepatic metabolism, (iii) high variability in PK parameters, (iv) food-related decrease in bioavailability, and (v) dose-related hepatotoxicity. Trigriluzole (BHV-4157) is a third generation prodrug of riluzole with improved PK/PD properties. In our MASS20 in vivo murine model of melanoma, GRM1 signal transduction appears to suppress tumor immunity through up-regulation of M-CSF and CCL2 expression with a subsequent increase in the percentage of M2 macrophages in the tumor microenvironment. In an immunocompetent mouse model, treatment with trigriluzole and anti-PD-1 antibody was more effective than either drug alone. Methods: Subjects with advanced or refractory solid cancers or lymphoma will be treated with increasing doses of trigriluzole in this phase 1b study. The dosing cohorts of trigriluzole will be using a semi-Bayesian modified toxicity probability interval dose escalation procedure. Trigriluzole monotherapy will be given for a 14-day lead in period and then patients will receive trigriluzole in combination with nivolumab 240mg IV every 2 weeks. After the maximum tolerated dose (MTD) of trigriluzole is identified, it will be checked in a cohort of 6 patients in combination with pembrolizumab. Total sample size will be 12 – 27 subjects. PD-1-directed therapy prior to entry is allowed. Blood samples and optional biopsies will be collected for correlative analysis. PBMCs and fresh tumor tissue will be grafted into NSG mice to create a patient-derived autologous double-humanized murine model for each participant. Clinical trial information: NCT03229278.

Background: Coxsackievirus A21 (CVA21) is a novel bio-selected oncolytic, immunotherapeutic agent. Intratumoral (i.t.) CVA21 injection can induce selective tumor-cell infection, immune-cell infiltration, IFN-γ response gene up-regulation, increased PD-L1 expression, tumor cell lysis and systemic anti-tumor immune responses. Preclinical studies in an immune-competent mouse model of melanoma have revealed that combinations of i.t. CVA21 and anti-PD-1 blockade mediate significantly greater antitumor activity compared to use of either agent alone. A clinical trial evaluating combination CVA21 and pembrolizumab in patients with melanoma was initiated and preliminary data on a pre-established futility endpoint are presented here. Materials and Methods: This is a single-arm, multi-institutional open-label phase Ib clinical trial of i.t. CVA21 and i.v. pembrolizumab for treated or untreated unresectable Stage IIIC-IVM1c melanoma. Subjects with injectable disease receive up to 3 x 108 TCID50 CVA21 i.t. on Days 1, 3, 5, 8, and then every 3 weeks for up to 19 injections. Subjects also receive pembrolizumab (2mg/kg) i.v. every 3 weeks starting on Day 8. The primary endpoint is safety/tolerability by incidence of dose-limiting toxicity. Secondary endpoints include best ORR by immune-related response criteria, progression-free survival, overall survival, quality of life, changes in melanoma-specific T cells, PD-L1 expression and Th1/Th2 gene expression profiles. The protocol included a futility analysis after the first 12 patients. Results: To date, 14 subjects have started on protocol therapy. Overall, the adverse events have been low-grade constitutional symptoms related to CVA21 and expected pembrolizumab-related side effects. No DLT’s have been reported. Currently, 11 patients are evaluable for investigator response assessment, not including 2 subjects who have not yet reached their first assessment and 1 subject who left the study early due to an unrelated adverse event. Among the evaluable subjects, the ORR was 73% (8/11). The DCR (CR+PR+SD) is currently 91% (10/11). In subjects with stage IVM1c disease, the ORR and the DCR is 100% (5/5). The study has met its primary statistical futility endpoint of achieving ≥2 confirmed objective responses (CR or PR) in the first 12 patients enrolled. Currently, the median time to response is 1.6 months. One of the 8 responders displayed early pseudo-progression and later developed a partial response. Conclusions: At a pre-specified futility analysis, combination CVA21 and pembrolizumab appears to be well-tolerated. Early tumor monitoring has identified encouraging reductions in a number of injected and non-injected lesions. Based on these initial results, the sample size has now been expanded to enroll up to 50 patients. Combination therapy of CVA21 and pembrolizumab may represent a new approach for the treatment of patients with injectable advanced melanoma. Citation Format: Ann W. Silk, Howard Kaufman, Nashat Gabrail, Janice Mehnert, Jennifer Bryan, Jacqueline Norrell, Daniel Medina, Praveen Bommareddy, Darren Shafren, Mark Grose, Andrew Zloza. Phase 1b study of intratumoral Coxsackievirus A21 (CVA21) and systemic pembrolizumab in advanced melanoma patients: Interim results of the CAPRA clinical trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr CT026. doi:10.1158/1538-7445.AM2017-CT026

6617 Background: Immuno-oncology (I/0) agents represent an important, accelerated breakthrough in cancer therapy. Like other MABs, these agents have been adjusted after FDA approval to have a flat-dose. Purportedly, flat dosing simplifies prescribing, dispensing, inventory and billing. Nivolumab (N) and Pembrolizumab (P) achieved FDA approval in several malignancies. Original studies determined the dosing of N @ 3 mg/kg q2-weeks and P @ 2 mg/kg q3-weeks. Based on simulations from population pharmacokinetics models, the FDA approved a FD of N 240 mg in melanoma, RCC and NSCLC. The financial impact of this FD methodology in our patient population was compared with weight based dosing (3mg/kg) of N with a cap of 240 mg versus flat dosing. The potential impact of this change was also evaluated for P, including if a 50 mg vial of P was still available. Methods: Applicable dispensed doses (N & P) and patients’ weights for 2016 were mined from the electronic health record. Wholesale Acquisition Costs at end of year were used for financial comparison. Results: see table. Conclusions: Weight based dosing with a cap (N 240mg; P 200mg) versus flat dosing would have saved $198,567 and $80,037, respectively. Additionally, $760,351 would have been saved if 50 mg vials of P were available. With the current drug pricing structure, wide-scale adoption of flat dosing for I/O mABs may result in higher drug costs. Labeling I/O mABs with both weight-based and FD options and ensuring the availability of proper vial sizes, particularly multi-dose vials, to fit the population and dosing schema would restrain the costs of care. [Table: see text]

2586 Background: ONC201 is an orally active, small molecule selective antagonist of the G protein-coupled receptor DRD2 that has established a new class of compounds referred to as imipridones. A first-in-human trial of ONC201 defined its recommended phase II dose (RP2D) as 625mg using once every three week administration that was very well tolerated at doses that yielded antitumor effects. ONC201 also showed stimulatory effects on immune cells in preclinical studies, including increased intratumoral NK cell infiltration in xenografts. Based on the exceptional safety profile of ONC201, weekly dosing has been evaluated. Methods: This open-label, 3+3 dose-escalation study used a starting dose of 375mg and escalated to 625mg using a weekly administration schedule. The primary endpoint was to determine the RP2D of ONC201 and secondary endpoints included PD, PK, toxicity, and anti-tumor efficacy. Based on signs of clinical activity and preclinical tumor type sensitivity studies of ONC201, the patient population was enriched for advanced glioblastoma, prostate cancer, and endometrial cancer. Six additional patients were treated at the weekly RP2D. Results: The RP2D for the weekly regimen was defined as 625 mg. Twelve evaluable patients were treated at this dose level and no drug-related AEs > grade 1 occurred. Five patients had stable disease by RECIST criteria for 21-29 weeks. A metastatic prostate cancer who received 375mg ONC201 weekly had significantly diminished intensity in bone scans after 6 doses. PK was consistent with previous reports: Cmax consistently reaches therapeutic micromolar plasma concentrations, ~11 hour half-life, evidence of sustained and delayed activity, no systemic accumulation. In agreement with preclinical observations of ONC201-induced NK cell populations, a 2-10 fold increase in circulating activated NK cells was observed in 5 prostate cancer patients. Conclusions: ONC201 is well tolerated at an oral dose of 625mg weekly, exhibits sustained and late anti-cancer activity, and increases circulating NK cells. Observation in this study, and other clinical studies, warrant further evaluation of the immune oncology effects of ONC201. Clinical trial information: NCT02250781.

9566 Background: Ipilimumab-related colitis has been well described, but the clinical presentation of colitis related to MONO or COMBO therapy has not. We aimed to characterize clinical features that define this serious adverse event. Methods: This retrospective study included melanoma pts with colitis screened from 6 academic centers treated with either MONO or COMBO therapy. Clinically relevant colitis was defined as diarrhea/colitis symptoms requiring systemic steroids. The onset, management, and outcomes related to colitis were summarized. Results: We screened 866 pts and identified 72 with clinically relevant colitis. For the MONO cohort, the incidence was 3.5% (23/657) and 23.4% (49/209) in the COMBO cohort. MONO-colitis occurred at a median 33 weeks (wks) into therapy, while median onset was 8 wks with COMBO therapy. One pt had grade 5 colitis from COMBO therapy. Despite the use of systemic steroids, COMBO-colitis needed more infliximab therapy (38.8% vs 26.1%). The median prednisone equivalent dose was higher for COMBO therapy (1.5 vs 1 mg/kg), and the median taper was shorter for MONO-colitis (4 vs 6 wks). Steroid dose-escalation for worsening symptoms during taper was more common with COMBO-colitis (42.9% vs 17.3%). Relapse was similar between cohorts (MONO: 26.1%, COMBO: 20.4%). Relapses occurred more frequently in COMBO therapy with tapers shorter than the median (38.1% vs 25.0%), and more frequently in MONO therapy with steroid doses lower than the median (60% vs 21.7%). MONO-colitis pts (17.4%) were less likely than COMBO-colitis pts (46.9%) to be restarted on PD-1 therapy. When restarted, only 13% with COMBO therapy relapsed as compared to 50% with MONO therapy. Objective response rates for MONO and COMBO cohorts were 72.7% and 56.1%, respectively. Conclusions: Colitis occurred with a much higher incidence in COMBO therapy. MONO-colitis was associated with a milder course with later onset, shorter duration and lower dose of steroids, fewer dose-escalations, and need for infliximab as compared to COMBO therapy. Relapse of colitis was generally associated with shorter steroid tapers for COMBO therapy and lower steroid doses for MONO therapy.

35 Background: A high mutation burden a biomarker of response to immune checkpoint therapy in melanoma, non-small cell lung cancer, and colorectal cancer. It is unknown whether mutation burden is predictive of response in other cancer types, and whether it is identifiable using available clinical assays. Methods: Using data for 10,745 tumors from 33 solid cancer types in TCGA, we looked for a mutation burden threshold (iCAM) in each cancer type associated with gene expression signatures of a blocked immune response of robust CD8+T cell response and up-regulation of immune checkpoint genes. Results: A unique iCAM threshold was identified in nine cancers: melanoma and lung, colon, endometrial, and gastric adenocarcinoma; serous ovarian, bladder urothelial, cervical, and ER+ HER2− breast cancer. iCAM thresholds applied to published clinical data for patients treated with immune checkpoint therapy showed that iCAM+ patients had significantly better response. iCAM+ tumors can be identified from clinical-grade NGS assays with high accuracy . In a prospective melanoma cohort of patients treated with anti- PD-1 patients with iCAM+ tumors had significantly better objective response rates, progression free survival, and overall survival compared patients to iCAM− tumors. Pattern of somatic mutations were different in iCAM+ and iCAM− tumors, suggesting different mechanism of carcinogenesis in iCAM+ versus iCAM− tumors. Higher fractions of leukocytes were NK (Natural Killer) cells and macrophages were M1 polarized, and a lower fraction of T cells were regulatory T cells in iCAM+ tumors compared to iCAM− tumors. Over 75% of the genes significantly up-regulated in iCAM+ tumors in every cancer types were in the immune response pathway, confirming immune response to be the main differentiator in iCAM+ veruss iCAM− tumors. Conclusions: In nine cancer types, a clinically useful mutation burden threshold (iCAM) associated with gene expression signatures of blocked immune response can identify likely responders to immune checkpoint therapy, and iCAM status is identifiable using currently available clinical NGS assays.

122 Background: GRM1 transgenic mice in an immune competent background (C57BL6) develope melanoma with 100% penetrance (TG-3 mice). Ectopic expression of GRM1 in immortalized mouse melanocytes (MelanA) produce cells (MASS20) that are tumorigenic in both SCID and immune competent C57BL6 mice. We have examined the downstream events of GRM1 signaling that allow transformed cells to evade immune destruction. Methods: Cytokine antibody arrays and Western blot were used to examine differences in cytokine expression between MASS20 cells and parental controls. Next, we cultured C57BL6 bone marrow cells in MASS20 conditioned medium for 24 and 48 hours and determined resultant M2 macrophage levels. Finally, we used oral weekly liposome clodronate (LC) treatment to deplete macrophages in murine models to determine the effect on tumor growth. Results: MASS20 cells produce higher levels of M-CSF and CCL2 as compared with control cells. These two cytokines recruit and induce the polarization of tumor associated macrophages (TAM) to increase the M2 fraction resulting in accelerated tumor progression. Culturing C57BL6 bone marrow cells in MASS20 conditioned medium increased expression of the M2 macrophage marker arginase1 at both time points. Weekly oral LC treatment to C57BL6 mice greatly inhibited the growth of MASS20 tumors as compared to growth in control animals. Single-cell suspensions of spleen and tumor cells from these animals were subjected to flow cytometry and showed that LC treatment depleted macrophages by 90% in the spleen, and increased the percentage of cd8+T and cd335+ NK cells in tumors from treated mice as compared to controls. Finally, we treated TG-3 transgenic mice with weekly oral LC and found that treated animals still developed tumors, but progression was greatly slowed, and survival greatly prolonged, compared to untreated animals. Within 1 week of stopping LC in the treated animals there was rapid growth of the tumors. These findings suggest a link between GRM1-signal transduction and TAM-induced suppression of tumor-immunity in melanoma. Conclusions: We conclude that GRM1-induced suppression of tumor immunity occurs through induction of M2 macrophages in the tumor microenvironment.

2514Background: ONC201 is an orally active, first-in-class small molecule activator of the integrated stress response that selectively upregulates ATF4 to trigger tumor cell death. Preclinical studies have established the compelling antitumor efficacy of the small molecule in a range of advanced solid tumors as well as its excellent safety profile in animals. A first-in-human phase I study of ONC201 was conducted to determine its recommended phase II dose (RP2D) in patients with advanced solid tumors. Methods: This open-label study treated 10 patients during the single-patient cohort dose escalation portion of the trial. The human starting dose was 125mg, which represents 1/10th of the no-adverse-adverse-effect-level in animal toxicology studies using standard allometric conversion. Results: The RP2D was defined as 625 mg and no Grade > 1 drug-related adverse events occurred on this trial, which enrolled patients at increasing dose levels of 125mg, 250mg, 375mg, 500mg, and finally 625mg. Pharmacokinetic a...

Molecularly targeted agents have shown activity in soft tissue sarcoma (STS) and benign connective tissue tumors over the past ten years, but response rates differ by histologic subtype. The field of molecularly targeted agents in sarcoma is increasingly complex. Often, clinicians must rely on phase II data or even case series due to the rarity of these diseases. In subtypes with a clear role of specific factors in the pathophysiology of disease, such as giant cell tumor of the bone and diffuse-type tenosynovial giant cell tumor, it is reasonable to treat with newer targeted therapies, when available, in place of chemotherapy when systemic treatment is needed to control disease. In diseases without documented implication of a pathway in disease pathogenesis (e.g. soft tissue sarcoma and vascular endothelial growth factor), clear benefit from drug treatment should be established in randomized phase III trials before implementation into routine clinical practice. Histologic subtype will continue to emerge as a critical factor in treatment selection as we learn more about the molecular drivers of tumor growth and survival in different subtypes. Many of the drugs that have been recently developed affect tumor growth more than survival, therefore progression-free survival may be a more clinically relevant intermediate endpoint than objective response rate using Response Evaluation Criteria In Solid Tumors (RECIST) in early phase sarcoma trials. Because of the rarity of disease and increasing need for multidisciplinary management, patients with connective tissue tumors should be evaluated at a center with expertise in these diseases. Participation in clinical trials, when available, is highly encouraged.

Background: ONC201 is an orally active first-in-class small molecule with strong antitumor activity in preclinical models of advanced cancers. In cancer cell lines and patient samples ONC201 induces activation of the integrated stress response (Ishizawa et al ASH, 2014) resulting in upregulation of ATF4 and CHOP which in turn regulates several proapoptotic genes namely DR5. ONC201 also causes late-stage inactivation of Akt and ERK, which also results in downstream activation of the apoptotic TRAIL pathway as part of innate immune surveillance (Allen J et al, Sci Trans Med, 2013). Activity of ONC201 is independent of p53 status and mutation agnostic. ONC201 is well tolerated at efficacious doses in animal models, crosses the blood brain barrier, is particularly effective in refractory tumors, depletes cancer stem cells (Ishizawa et al, ASH 2014; Prabhu et al, Blood 2014; Zhao et al, ASCO 2014), and is effective with infrequent dosing preclinically. Based on the compelling efficacy and safety profile of ONC201 as well as the engagement of signaling pathways critical for many cancers, the clinical introduction ONC201 in advanced cancer patients is warranted. Methods: The first-in-human study of ONC201 (NCT02250781) began in January 2015 as an open-label single-site phase I trial enrolling adult patients with refractory advanced solid tumors and glioblastoma (GBM). Patients with symptomatic brain metastases or prior bevacizumab for treatment of GBM are excluded. The primary endpoint is determination of the recommended phase II dose (RP2D) of single agent ONC201 given orally once every 3 weeks (1 cycle). Secondary endpoints include assessment of pharmacodynamics using select biomarkers for ONC201 (Allen et al, 2015), pharmacokinetics, toxicity, and efficacy. The study employs an accelerated, single patient per cohort, dose escalation design with expansion to a standard 3+3 design if a subject has grade > / = grade 2 toxicity or dose limiting toxicity within cycle 1. The maximum tolerated dose is the highest dose level in which 6 patients have been treated with < / = 1 instance of DLT. Based on pre-clinical models doses for once every 3 week dosing were: dose leve1 (DL) 1- 125mg; 2- 250mg; 3- 375mg; 4- 500mg; 5- 626mg. Results: Single patient dose escalation was completed in cohorts 1 to 4 and 6 patients completed at least 3 weeks of level 5 without any > / = grade 2 drug related toxicity. At DL 6, Mean T1/2 = 7.91hours (7.01-9.42); Mean Cmax = 8.58 ug/mL (3.95 - 19.15) Mean AUC(0-24) = 28057 (15293 - 50597) hr-ng/mL. Ongoing radiographic stable disease is noted in a patient with metastatic castrate resistant prostate cancer. Conclusions: The RP2D of ONC201 is 625mg on a once every 3 week schedule. Evaluation of PD markers of response and enrollment to a dose expansion safety cohort are ongoing. Citation Format: Mark N. Stein, Rebecca Moss, Tina Mayer, Ann W. Silk, Nancy Chan, Ling Zheng, Yasmeen Beckett, Noelle Tenpenny, Edward Bentlyewski, Robert DiPaola, Rohinton Tarapore, Joshua Allen, Janice Mehnert. First-in-human dose escalation study of oral ONC201 in advanced solid tumors. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr C137.

Importance The treatment of multiple brain metastases (MBM) from melanoma is controversial and includes surgical resection, stereotactic radiosurgery (SRS), and whole-brain radiation therapy (WBRT). Several new classes of agents have revolutionized the treatment of metastatic melanoma, allowing some subsets of patients to have long-term survival. Given this, management of MBM from melanoma is continually evolving. Objective To review the current evidence regarding the treatment of MBM from melanoma. Evidence Review The PubMed database was searched using combinations of search terms and synonyms for melanoma, brain metastases, radiation, chemotherapy, immunotherapy, and targeted therapy published between January 1, 1995, and January 1, 2015. Articles were selected for inclusion on the basis of targeted keyword searches, manual review of bibliographies, and whether the article was a clinical trial, large observational study, or retrospective study focusing on melanoma brain metastases. Of 2243 articles initially identified, 110 were selected for full review. Of these, the most pertinent 73 articles were included. Findings Patients with newly diagnosed MBM can be treated with various modalities, either alone or in combination. Level 1 evidence supports the use of SRS alone, WBRT, and SRS with WBRT. Although the addition of WBRT to SRS improves the overall brain relapse rate, WBRT has no significant impact on overall survival and has detrimental neurocognitive outcomes. Cytotoxic chemotherapy has largely been ineffective; targeted therapies and immunotherapies have been reported to have high response rates and deserve further attention in larger clinical trials. Further studies are needed to fully evaluate the efficacy of these novel regimens in combination with radiation therapy. Conclusions and Relevance At this time, the standard management for patients with MBM from melanoma includes SRS, WBRT, or a combination of both. Emerging data exist to support the notion that SRS in combination with targeted therapies or immune therapy may obviate the need for WBRT; prospective studies are required to fully evaluate the efficacy of these novel regimens in combination with radiation therapy.

TPS2623 Background: ONC201 is an orally active first-in-class small molecule with strong antitumor activity in preclinical models of advanced cancers. In cancer cell lines and patient samples ONC20...

TPS9079 Background: Despite recent advances in the treatment of metastatic melanoma, survival remains poor in patients who develop brain metastases. Radiation therapy continues to be the backbone o...

TPS9092 Background: The incidence of SCCS has increased over the past two decades, including a high risk subset with aggressive behavior. Due to a lack of high-quality clinical trials in this popul...

Meeting abstracts The efficacy and safety of high dose IL-2 therapy in patients with brain metastases due to renal cell carcinoma is not well characterized. Data were prospectively collected in a registry of 371 patients with RCC receiving high-dose IL-2, including 18 patients with a history of