Immunotherapy Bridge 2018 and Melanoma Bridge 2018: meeting abstracts

© The Author(s) 2019. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat iveco mmons .org/licen ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creat iveco mmons .org/ publi cdoma in/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Immunotherapy Bridge 2018

Background: Stage IV melanoma has a 5-year survival rate of only 15-20%. Immune checkpoint blockade has shown therapeutic efficacy in a subset of melanoma patients, often those with pre-existing antitumor immunity. Therapeutic vaccines targeting melanoma-associated antigens are commonly immunogenic, but only rarely effective in promoting clinical responses, suggesting a clear need for further improvement. Materials and methods: To promote strong anti-tumor immune responses in melanoma patients, we created a vaccine consisting of autologous dendritic cells (DC) transduced ex vivo with a recombinant adenovirus encoding three shared melanoma antigens: Tyrosinase, MART-1, and MAGE-A6 (TMM2). Monocyte-derived DC were first matured with IFN-γ + LPS, and then transduced with recombinant adenovirus encoding TMM2 and administered to patients (n = 35) 3× via bi-weekly intradermal injections in a Phase I trial. Human microarrays were used to analyze gene expression profiles of patient immature, mature, and adenovirus-transduced DC. Results: Genomic analyses revealed that melanoma patient (but not healthy donor [1]) DC exhibit a significant increase in expression of transcripts encoding immunosuppressive molecules, such as IDO1 and TXN after ex vivo maturation and viral transduction, when compared to individual-matched immature DC. DC-associated transcripts correlating significantly with clinical outcome include LAMP1, DDO, CLEC4A and NLRP2 linked to antigen-presentation, aspartate degradation, plasmacytoid DC and an inhibitor of TBK1/Type-1 IFN signaling, respectively. Protein-based analyses indicate ICOSL downregulation on the surface of immature, mature, and transduced DC in melanoma patients versus HD. NFkB signaling, known to regulate ICOSL expression in DC, appears selectively dysregulated in mDC generated from melanoma patients versus HD. Conclusions: Our profiling data suggest deficiencies in NFkB and Type-1 IFN signaling and costimulatory molecule expression (ICOS-L), as well as elevated expression of immunosuppressive gene products (IDO1, TXN), may serve to limit the immunostimulatory capacity of melanoma patient-derived DC and derivative DC-based vaccines. These data provide clues for targeted manipulation of patient DC to develop improved vaccines implementing DC for the treatment of advanced-stage melanoma patients. Trial Registration Identifier NCT01622933. Background: Anti-PD-1 plus anti-LAG3 therapy was well tolerated in melanoma patients and has shown clinical activity in patients progressive on PD-1 therapy. However, a significant proportion of patients still exhibit a priori resistance or suffer from disease progression. Domatinostat, a selective HDAC class I inhibitor, has been preclinically reported to upregulate MHC molecules, tumor-associated antigen expression, and increase inflammatory signature and T cell infiltration into tumors. Here, we provide preclinical data on the triple combination of domatinostat, anti-PD-1 plus anti-LAG3 antibodies. Materials and methods: Anti-tumoral efficacy and the impact on tumor microenvironment were analyzed in a syngeneic C38 model with a low response rate to anti-PD-1 and no response to anti-LAG3 treatment, reflecting refractory clinical situation. Clinical equivalent dose of domatinostat was used to ensure translational relevance. Results: Domatinostat and anti-PD1 in monotherapy reduced C38 tumor growth resulting in 10% and 25% regressions, respectively. The combination of domatinostat plus anti-PD-1 antibody resulted in a stronger tumor growth control with 60% regressing tumors. Anti-LAG3 treatment alone or in combination with either domatinostat or anti-PD-1 did not show added anti-tumoral efficacy. However, addition of domatinostat to anti-PD-1+ anti-LAG3 double combination resulted in a higher anti-tumor activity with a tumor regression rate of 80%. FACS analysis revealed that domatinostat increased expression of MHC class I molecules on tumor cells, and MHC class II molecules on tumor cells, Ly6G+ myeloid derived suppressor cells and M1 macrophages. Domatinostat-mediated upregulation of MHC class II and co-stimulatory molecules [Bretz et al. AACR 2017] allows to hypothesize that the addition of an anti-LAG3 antibody, which blocks detrimental effect of MHC class II/LAG3 engagement on T cell activity, may beneficially affect T cell response by supporting T helper cell activation and function. Consistently, in combination with both checkpoint inhibitors domatinostat strongly enhanced the proportion of proliferating cytotoxic T cells (CTLs) in the tumors and reduced exhaustion phenotype of CTLs and T helper cells.
Conclusions: Domatinostat has previously demonstrated beneficial immunomodulatory effects in syngeneic preclinical tumor models in combination with immunostimulating agents as well as with immune checkpoint blockade. Currently, domatinostat is in clinical evaluation in combination with pembrolizumab in advanced melanoma patients refractory or non-responding to anti-PD-1 ("SENSITIZE" study; NCT03278665). The triple combination of domatinostat, anti-PD-1 and Background: Epidermal growth factor receptor (EGFR) plays an important role in cancer progression. However, the impact of anti-EGFR therapies on patients overall survival has been limited mainly by the emergence of different forms of tumor resistance [1]. N-glycolyl variant of GM3 ganglioside (NGcGM3) is specific antigen expressed in some tumors, it has received some attention as a privileged target for cancer therapy [2]. This ganglioside has been associated with a poor prognostic in colon and lung cancer [3,4]. Several reports have documented a functional relationship between GM3 and EGFR at tumor cell membrane. GM3 physically associates to EGFR inhibiting its ligand depend phosphorylation, but it also facilitates an alternative/compensatory signaling cascade mediated by Uroquinase Plasminogen Activator Receptor (uPAR) and integrin α5β1 interaction with membrane reduction of NAcGM3 ganglioside [5]. However, no definite information happens on the difference between N-glycolyl and N-Acetyl variants of GM3 regarding this interaction [6].

Materials and methods:
The impact study of a combination of immunotherapies against these two "physically and/or functional related targets NGcGM3 and EGFR" were performed in two murine lung metastasis models: (Lewis lung carcinoma (3LL-D122) in C57BL/6 and mammary carcinoma (4T1) in BALB/c). 7A7 murine monoclonal antibody (mMAb) against murine EGFR was the therapy selected to this molecule [7]. While, two therapies against NGcGM3 ganglioside were evaluated NGcGM3/VSSP vaccine (NeuGcGM3 was incorporated in the outer membrane protein complex of Neisseria meningitidis bacteria to form very small size proteoliposomes (VSSP)) [8]  Infliximab was the most used therapy (fifteen cases) followed by mycophenolate mofetil (six cases). Methotrexate and IVIG were used in two cases each. Another subsequent therapy was required in three cases and four immunosuppressive treatments in one case. In nineteen cases a complete recovery of the irAEs was observed. The median time to resolution was 11 weeks (min: 1; max: 40). Conclusions: For a minority of patients, solely therapy with corticosteroids is not sufficient. The interdisciplinary exchange about these toxicities is crucial, since experiences of individual therapy courses can be adapted to other patients.
Background: Immunotherapy and targeted therapies have improved the prognosis of patients with metastatic melanoma [1,2,3]. The objective of this study was to evaluate (1) the efficacy and safety of retreatment with nivolumab and (2) that of concurrent immune checkpoint inhibitor therapy and radiotherapy (immunoradiotherapy) in patients with metastatic melanoma after progression on nivolumab. Patients and methods: (1) A retrospective review was performed on eight consecutive metastatic melanoma patients retreated with nivolumab who progressed on previous nivolumab. These patients received nivolumab 2 mg/kg every 3 weeks. Best responses to each treatment were assessed using RECIST 1.1. (2) A retrospective review was performed on 16 consecutive patients with metastatic melanoma treated with concurrent immunoradiotherapy after progression on nivolumab. Best responses to immunoradiotherapy were assessed either inside or outside of the radiation fields. The target lesions ratio (the sum of the diameters of the target lesions inside the irradiated fields/all target lesions) was also assessed. Results: (1) Three out of eight patients received chemotherapy before first nivolumab. The median first nivolumab treatment period was 4.1 months. Three (37.5%) patients achieved a partial response and three (37.5%) patients achieved stable disease as their best response. Between first and second nivolumab, patients were treated with ipilimumab (n = 6), vemurafenib (n = 1), or no other medical treatment (n = 1). Four patients received radiation therapy. The median second nivolumab treatment period was 4.3 months. Two (25%) patients who received second nivolimab achieved a partial response and three (37.5%) patients achieved stable disease as their best response. Among the four patients treated with ipilimumab and radiotherapy between first and second nivolumab, the response rate was 50% and the disease control rate was 75%. (2) Among the patients, seven received ipilimumab and radiotherapy (Ipi-RT), six received nivolumab and radiotherapy (Nivo-RT), and three sequentially received Ipi-RT and Nivo-RT. As shown in Table 1, the overall response rate (all patients regardless of inside or outside radiation fields) was 30%. The response rate inside the radiation fields was 68.8% for all patients combined.
The response rates of Ipi-RT and Nivo-RT inside the radiation fields were 37.5 and 100% (P = 0.03), respectively. Grade 3 adverse events were observed in three patients treated with Ipi-RT. The target lesions ratio was a predictive marker of disease control rate among patients treated with Nivo-RT.

Conclusions:
This study showed that retreatment with nivolumab or concurrent immunoradiotherapy is an option for patients with metastatic melanoma after progression on nivolumab. Consent to publish: Informed consent including consent to publish was obtained from all individual participants included in the study. Comparing the dispensed drug in drug day to the drug really bought we saved 4993 mg (approximately 50 vials of 100 mg) with an economy of €59,058; this is related to the overfill of injecting drug vials, corresponding to about 10-12 mg over the declared amount of drug, as prescribed by F.U. XII Edition (2.9.17) and by FDA guidelines on the filling volume in excess of vials. Conclusions: Nivolumab flat dose recently introduced changed not only clinical practice but also therapy costs in particular the doses were increased so the costs were higher. However drug day allows to recover all drug residues, so it is a very effective tool for the containment of pharmaceutical costs.
Background: Merkel cell carcinoma (MCC) is a rare and aggressive neuroendocrine skin cancer which had an increasing incidence in the Swedish population (1). The pathogenesis is linked to the immune system and immunocompromised patients have an increased risk of developing MCC (2). With the discovery of Merkel cell polyoma virus (MCPyV), which is present in about 80% of MCC (3) the suggestion of a connection with the immune system has been strengthened. Previous reports (4) has shown that patients with a virus-negative disease have a worse prognosis. Furthermore, recent treatment with PD1 or PDL-1 blockade has shown to be successful in patients with advanced MCC (5,6). Our aim is to analyze clinical variables and their prognostic impact in a Swedish cohort of MCC-patients, in order to better understand possibilities to personalize treatment in the future. Methods: 106 MCC patients referred to the plastic surgery unit at Karolinska University Hospital from 1989 to 2017 were included and retrospective data was collected. 65 (61%) were identified as female, 41 (39%) as male. Mean age at operation was 75.5 years, mean overall survival (OS) was 4.2 ± 0.5 years. MCPyV status was available in 42 patients, 5 of these were excluded due to lacking retrospective data. Out of the 37 patients, 24 (65%) patients were MCPyV positive, 13 (35%) MCPyV negative. Results: Survival analysis did not show any significant difference between MCPyV positive and negative patients. When gender was added into multivariate analysis we found that female patients with virus negative disease had a significantly better outcome than virus negative male patients (p = 0.02) whereas virus positive MCC did not show any significant difference in OS between female and male patients.
Conclusions: Female patients with negative MCPyV MCC had a better outcome than the male patients. This finding indicates that MCPyV positive and negative MCC act as two different diseases. It also raises questions if there is a difference in the disease or immune response between male and female patients.    The irAEs were associated with ipilimumab + nivolumab in 60 cases, with pembrolizumab in 46 cases, with nivolumab in 18 cases and with ipilimumab in 16 cases. More detailed information is available in Table 1. Hospitalization was required for almost half of the cases and the majority of the irAEs persisted for more than 4 weeks (50%). Corticotherapy was used in 55% of the irAEs. Excluding endocrine irAEs, which didn't recover completely, a full recovery was observed in 43% of the cases. Conclusion: Severe irAEs are frequent in patients treated with IT in the daily practice. These can occur as soon as 1 week after initiating therapy but also much later-up to 3 years in our collective. This highlights the need to thoroughly inform this population before, during and after receiving IT. Although most of the irAEs responded to steroid administration and temporary treatment suspension, 50% lasted for more than 4 weeks and hospitalization was required in a high percentage of cases, increasing the costs associated with AE management.   To date, the anatomic extent of tumor (TNM-classification) has been by far the most important factor to evaluate the prognosis of cancer patients. However, this classification provides limited prognostic information and does not predict response to therapy. We redefined cancer by integrating the immune system, to transfer cutting-edge medicine to the patients. We have previously shown that tumors from human colorectal cancer with a high-density of infiltrating memory and effector-memory T-cells (Tem) are less likely to disseminate to lymphovascular and perineural structures and to regional lymph-nodes. We also demonstrated the critical tumor-microenvironment parameters determining the dissemination to distant metastasis. We found that the combination of immune parameters associating the nature, the density, the functional immune orientation and the location of immune cells within the tumor was essential to accurately define the impact of the local host-immune reaction on patients' prognosis. We defined these parameters as the "immune contexture". We characterized the immune landscape within human tumors, and showed the importance of adaptive immune cells including, cytotoxic T-cells, Th1-cells, B-cells and T-follicular-helper (Tfh) cells. We described the immunophenotype and antigenome associated with immune escape mechanisms and demonstrated mechanisms associated with preexisting and proliferating intratumoral T-cells. Based on the immune contexture, a standardized, simple and powerful digital-pathology-based immune stratification-system, termed "Immunoscore", was delineated having a prognostic power superior to that of the currently used cancer staging-system. Tumor invasion parameters were statistically dependent on the host-immune reaction. A worldwide consortium validated the prognostic value of the consensus Immunoscore, using a standardized-assay. Very recently, we conducted comprehensive analyses revealing a large inter and intra-metastatic immune heterogeneity. Nonetheless, even when measured on a single biopsy, the Immunoscore held a prognostic value and was surpassing the accuracy of PD-L1 expression. Our most recent data supported the significant role of Immunoscore and immunoediting in affecting metastatic dissemination. We hence proposed a "parallel selection model" of tumor evolution incorporating the effects of the immune system in shaping and driving metastatic spread. Thus, tumor progression, invasion and recurrence are dependent on preexisting immunity and on Immunoscore.   Background: With the increasing use of anti-PD-1 in patients with melanoma and other tumor types, there is great interest in developing an early on-treatment biomarker that correlates with long-term patient outcomes. An understanding of the pathologic features of immune-mediated tumor regression is key in this endeavor. The objective of this study was to identify the features of anti-PD-1-associated immune-mediated melanoma regression and correlate them with patient outcomes.

Materials and methods:
We assessed hematoxylin and eosin (H&E)stained slides, from formalin-fixed paraffin-embedded (FFPE) tissue, for pathologic features in N = 139 pre-and on-treatment melanoma specimens from 79 unique patients treated with anti-PD-1. Response to therapy was assessed using RECIST v1.1. The discovery cohort consisted of archival specimens (n = 30) from patients with advanced melanoma treated at Johns Hopkins; validation cohort specimens (n = 109) originated from patients enrolled on the nivolumab monotherapy arm of CA209-038 (NCT01621490). Immune and non-immune pathologic features such as lymphoid aggregates, neovascularization, plasma cells, proliferative fibrosis, necrosis, and dense collections of melanophages were studied. Tumor infiltrating lymphocyte (TIL) density was also scored. Pathologic features were compared between preand on-anti-PD-1 therapy specimens and between responders and non-responders, and a constellation of immune-related pathologic response (irPR) features was identified. These were used to develop an irPR score (from 0 = no irPR features to 3 = major pathologic response on biopsy (MPR bx , ≤ 10% residual viable tumor). In the validation cohort, irPR scores were evaluated for association with patient outcome.

Conclusions:
We have identified histopathologic features of immunemediated tumor regression on routine H&E-stained slides in patients with melanoma that correlate with objective response to anti-PD-1 and OS. This approach is inexpensive and widely available. Our findings inform grading systems for pathologic response following administration of checkpoint blocking agents and will likely influence future biomarker strategies.
Background: Although much clinical progress has been made in harnessing the immune system to recognize and target cancer, there is still a significant lack of an understanding of how tumors evade immune recognition and the mechanisms that drive tumor resistance to both T cell and checkpoint blockade immunotherapy. Our objective is to understand how tumor-mediated signaling through multiple inhibitory receptors, including PD-1, combine to affect the process of T cell recognition of tumor antigen and activation signaling. This with the goal of understanding the basis of resistance to PD-1 blockade and potentially identify new molecular targets to enable T cells to overcome dysfunction mediated by multiple inhibitory receptors. Methods and Results: Biomembrane Force Probe (BFP) measurements show that that the activities of TCR-proximal signaling components affect T cell mechanosensing and sensitivity at the earliest stages of antigen recognition and are influenced by PD-1 and other inhibitory receptors via Shp-1/2 by targeting CD28 and Lck to directly suppress TCR-pMHC-CD8 binding. Phospho-proteomics and flow cytometry-based analysis of patient-derived T cells from PD-1 responders and non-responders identified additional mediators, signaling components and pathways associated with PD-1 checkpoint blockade resistance. CRISPR/Cas9 mediated genome editing was utilized to determine if resistance is mediated by the continued signaling of multiple IRs by perturbing IR signaling in mouse models of PD-1 blockade.

Conclusions:
Our results suggest that T cell activation signaling and effector function is altered in PD-1 blockade-resistant patients due to the combined signaling of multiple inhibitory receptors, and therefore clinical targeting of pathways common to multiple inhibitory receptors can overcome PD-1 blockade resistance. Targeting these interactions and understanding the basis of resistance to PD-1 blockade would potentially allow identification of novel biomarkers of resistance or new molecular targets to enable T cells to overcome dysfunction during PD-1 checkpoint blockade and improve patient outcomes. J Transl Med 2019, 17(Suppl 1):16 The combination of tavokinogene telseplasmid (plasmid IL-12 or "tavo") and pembrolizumab has been shown to induce frequent, durable clinical responses in patients with immunologically "cold" melanoma. To assess the unique contribution of intratumoral (IT) tavo to systemic anti-tumor immune responses, we examined regression of untreated lesions, the hallmark of systemic response to IT immunotherapy, in patients treated with IT-tavo monotherapy. Stage III/IV melanoma patients were enrolled in a phase 2 trial where 1-4 accessible lesions were treated and at least one lesion was left untreated (NCT01502293). The sums of the diameters of treated and untreated lesions were assessed by modified RECIST (allows tumors > 0.3 cm and more lesions/organ). AEs were assessed by CTCAE version 4. Paired biopsies were assessed for PD-L1 levels by IHC and for changes in inflammatory gene expression. 51 patients were enrolled. 35 patients (69%) had been previously treated with systemic therapy (22 with 2 or more therapies). The most common AEs were pain and local reactions at treatment sites. There were no grade 4 or 5 AEs during the study. Two patients (3.9%, 2/51) had AEs leading to study withdrawal. Best overall response was 29.2% (10.4% CR [5/48 patients] and 18.8% PR [9/48 patients]). Time to objective response (min-max) was 0.9-3.9 months. The median duration of response was 7.3 months and the median time to progression was 2.7 months. The median OS was not met at median follow-up of 28 months. Longitudinal response assessment suggested 3 response categories: durable, transient, and no response. This clinical data demonstrates that IT-tavo can trigger potent systemic immune activation, evidenced by regression in 40% patients (16% of all untreated lesions) which dovetails with our previous paired tumor analysis highlighting that IT-tavo-EP increases TILs (both CD8 + T cells and NK cells) as well as intratumoral expression of interferon gamma associated genes. Our current data suggests that while this treatment-related abscopal effect and related increase of intratumoral inflammation can have a systemic impact on tumor burden, it also triggers adaptive resistance, manifested as increased PD-L1 expression, a major factor that could limit the clinical benefits of IT-tavo-EP monotherapy in some patients. This provides a strong rationale for combination with an anti-PD-1 therapy. In the context of our previously reported findings, our data also supports the hypothesis that anti-PD-1 Ab therapy amplifies this newly inflamed tumor microenviroment leading to durable, systemic responses in patients who would not otherwise benefit from PD-1 Ab monotherapy. Background: Immune checkpoint inhibition, such as anti-CTLA-4 and anti-PD-1 blockade, significantly improve survival of melanoma patients, however, a substantial proportion of patients do not respond or develop resistance. As tumor angiogenesis has been reported to be associated with reduced anti-tumor immunity, we reasoned that a kinase insert domain receptor (KDR) germline variant (Q472H) is a driver of melanoma angiogenesis and might determine the clinical response to anti-CTLA-4 and/or anti-PD-1 therapy. Methods: Germline DNA from 1429 stages I-IV melanoma patients enrolled in the NYU Interdisciplinary Melanoma Cooperative Group (IMCG) between 2010 and 2017 was genotyped to determine KDR variant status (MassARRAY iPLEX platform, Agena Bioscience). Clinical response was assessed for a cohort of melanoma patients (n = 206) treated with anti-CTLA-4 or anti-PD-1. Tumor angiogenesis was assessed by immunostaining with anti-CD34 antibody (Abcam) and quantitation of microvessel density (MVD) in melanoma tissues (n = 161). Results: The KDR Q472 variant allele was found in 37% (536/1429) of melanoma patients, which is significantly higher (p = 0.001) than its frequency of 21% in the general population and 13% in Caucasians (1000 Genomes Project database [http://www.inter natio nalge nome. org/]). Melanoma patients harboring germline KDR Q472H had significantly higher MVD (p = 0.04). Presence of KDR Q472H was not associated with response to anti-CTLA-4 (p = 0.8), but was associated with resistance to anti-PD-1 treatment (p = 0.02). Conclusions: Our results suggest that the germline variant KDR Q472H is enriched in melanoma patients compared to general population, and promotes an angiogenic tumor phenotype, which might contribute in part to resistance to anti-PD-1 treatment in the metastatic setting. The lack of association with anti-CTLA-4 treatment might be due to the different mechanisms of action as CTLA-4 is thought to regulate T-cell proliferation early in the immune response, mostly in lymph nodes, whereas PD-1 suppresses T cells later in the immune response, primarily in peripheral tumor tissue where angiogenesis might have more of an effect.

A KDR germline variant is associated with increased risk of melanoma, a pro-angiogenic phenotype and resistance to immunotherapy
Efficacy of novel melanoma treatments in metastatic melanoma patients with germline CDKN2A mutations. Background: Somatic mutations and deletions in the CDKN2A gene are frequent driver events in melanoma tumors. Germline CDKN2A mutation is one of the strongest known risk factors for cutaneous melanoma. Individuals that carry inherited CDKN2A mutations are extremely prone to develop melanomas and mutation carriers are also reported to have inferior melanoma-specific survival. Methods: CDKN2A mutation carriers with metastatic melanoma undergoing BRAF ± MEK or immune checkpoint inhibitors were included in the study and therapy responses assessed. From four publicly available datasets, melanomas with somatic CDKN2A mutation were analyzed for association with tumor mutational load. Results: Nineteen CDKN2A mutation carriers received BRAF ± MEK inhibitors, thirteen (68%) responded to the therapy, all with a partial response (no complete response) and this was not significantly different from an expected response rate of 64% (13% for complete responses), estimated from the phase III trials involving Dabrafenib, Vemurafenib, Dabrafenib/Trametinib and Encorafenib/Binimetinib 1-3. Nineteen carriers received checkpoint inhibitors, eleven (58%) responded to the therapy which was a significantly higher rate than expected (P = 0.03, binomial test against an expected rate of 37%, estimated from the ipilimumab, pembrolizumab, nivolumab and ipilimumab/nivolumab trials 4-7). Further, six of the nineteen carriers (32%) had complete response which was also a significantly higher rate than expected (P = 0.01, binomial test against an expected rate of J Transl Med 2019, 17(Suppl 1):16 7%). A significantly higher frequency of the CDKN2A mutation carriers receiving immunotherapy had M1c-d disease (79%), brain metastasis (26%) or were previously treated (63%), compared to the patients in the clinical trials. In a separate analysis involving mutation data from 879 tumor samples, the 118 melanomas harboring somatic CDKN2A mutations had significantly higher total numbers of mutations compared to 761 melanomas without CDKN2A mutation (Wilcoxon test, P < 0.001). Conclusions: The response to BRAF ± MEK inhibitors in the germline CDKN2A mutated melanoma patients was not significantly different from an expected response rate. However, the CDKN2A mutated melanoma patients had superior immunotherapy responses and this could be due to increased tumor mutational load in CDKN2A mutated tumors, resulting in more neoantigens and stronger antitumoral immune responses. These findings are primarily reassuring for CDKN2A mutation carriers that have a very high life-time risk to develop melanoma(s), the better than expected response to immunotherapy in the metastatic setting indicates that immunotherapy also could have a significant role in the adjuvant situation among such carriers. Background: A high proportion of skin cancer patients (e.g. advanced melanoma or merkel cell carcinoma patients) are refractory, do not respond to or relapse on checkpoint inhibition alone, therefore a high unmet medical need remains. One promising approach is to enhance immunogenicity and alter the tumor microenvironment to a more immune-inflamed phenotype by epigenetic intervention. Preclinical experiments suggest various immune-modulatory capabilities of the orally available selective class I HDAC inhibitor domatinostat rendering it as a favorable combination partner for different immunotherapy approaches. Our results provide the rationale and basis for domatinostat as key component for future immune-oncology combination approaches. This concept is currently tested in a Phase Ib multi-center study in advanced melanoma ('SENSITIZE'; NCT03278665). The SENSI-TIZE study tests the combination of domatinostat and pembrolizumab in patients not responding to prior immune checkpoint therapy for safety and tolerability, favorable modulation of the tumor microenvironment and clinical efficacy. Methods: Immunomodulatory effects of domatinostat were tested in cell-based assays and various syngenic mouse models. Changes in domatinostat-induced effects on gene expression and direct antitumor effects alone or in combination with different immunotherapy approaches were monitored. In SENSITIZE, advanced cutaneous melanoma patients which did not respond to prior checkpoint inhibitor treatment, domatinostat is combined with pembrolizumab to identify the optimal dose for the combination. Tumor assessments are conducted every 12 weeks and sequential tumor biopsies are taken for comprehensive analyses of immune cell infiltration into the tumor. Blood samples are collected in parallel to investigate PK, PD, and changes in gene expression profiles upon treatment. Results: Preclinical characterization demonstrates that domatinostat can modulate the tumor microenvironment leading to an 'inflamed' gene expression profile including upregulation of genes essential for antigen presentation and processing. Importantly, combination of domatinostat with different immunotherapy approaches results in synergistic anti-tumor effects. Based on this, the clinical study SEN-SITIZE has started and all patients for dose cohort 1 (100 mg domatinostat) have been enrolled and the safety review committee (SRC) has recommended the enrolment of patients for dose cohort 2 [200 mg]. Conclusions: Based on preclinical experiments suggesting that domatinostat can favorably modulate the tumor microenvironment making it more susceptible for combination for immunotherapy approaches in various indications including advanced melanoma. Our goal is to translate these findings to current SENSITIZE study in patients with advanced melanoma. Domatinostat in combination with pembrolizumab has been shown safe and well tolerated in the lowest dosing cohort (100 mg) and patients are now recruited into dose cohort 2 (200 mg). Biomarker and efficacy data will be taken into consideration to define optimal dosing of domatinostat in combination with anti-PD-1 antibodies. BRAF inhibitors have been a great success for patients with a BRAF V600 mutation, however, only about 50% of patients respond to single BRAF inhibitor therapy and the majority of these patients eventually relapse. We have established 53 melanoma cell cultures from biopsies naïve to BRAF inhibitor and progressive on BRAF inhibitor. We in vitro tested each cell line for resistance to BRAF inhibitor and found 27 to be resistant. Surprisingly, a few melanoma cultures from patients who have never been exposed to BRAF inhibitors had innate resistance, while the majority of cell cultures from progressive patients were resistant to in vitro BRAF inhibition. To elucidate the possible resistance mechanisms, we performed RNAseq on all 53 cultures and found differentially expressed genes in common with the innate and adaptive resistance melanoma cell cultures. The phenotype switching signature was one of the resistant mechanisms in common. Although we find some common signatures, there are other resistant cell cultures that have a unique gene signature, which suggests heterogeneity among resistance mechanisms.  advances, notwithstanding its unquestionable limitations of toxicity and high prices. Different approaches of active specific immunotherapy have been explored many years ago without success. Ganglioside vaccines are one them, with controversial results in the past. NGcGM3/ VSSP cancer vaccine was developed at the Center of Molecular Immunology, Havana; Cuba [1]. NeuGcGM3 ganglioside was incorporated in the outer membrane protein complex of Neisseria meningitidis bacteria to form very small size proteoliposomes (VSSP). Immunogenicity and low toxicity were evidenced in Preclinical and Clinical studies. NGcGM3 specific antibodies (IgM, IgG and IgA class) [2] and cellular (NK, CD8+ and CD4+ T cells) [3] responses were observed. Recently, inhibition mechanism of Src/FAK/Stat3 pathway was demonstrated [4]. Case reports: Clinical evidences of immunogenicity and antitumor response were associated with extended survival times in two patients treated subcutaneously with NGcGM3/VSSP cancer vaccine in a Phase Ib/IIa clinical trial [5]. Vaccine formulation was administered by five doses each 14 days (induction phase) and each 28 days (maintained phase) during 1 year as protocol design. Immunizations continued monthly for four additional years; after that, patients were vaccinated each 3 months. Previously, positive NGcGM3 ganglioside expression was tested in tumor biopsies with specific 14F7 monoclonal antibody (murine IgG1 against NGcGM3) [6]. One patient with lung metastases of cutaneous melanoma that refuses surgery, achieved disease stabilization with excellent quality of life for more than 13 years; another patient with intestinal metastases of amelanotic melanoma, in progressive disease after surgery, obtained complete response sustained for more than 10 years. Transient and reversible local symptoms were observed in both patients. Reversible hypersensibility reaction grade II was observed in the second patient. At the same time, humoral and cellular responses was measured, as signal of certain level of restoration of the patient's immunocompetence. Conclusions: NGcGM3/VSSP is an attractive target for active specific immunotherapy in melanoma patients, the rate of overall survival observed resembles those with Il-2 therapy. This antitumor responses of long duration with good performance status, constitutes an unusual clinical observation in the frame of a Phase Ib/IIa clinical trial developed. A new scenery for this cancer vaccine in metastatic melanoma patients might be in combination with anti-checkpoint therapy.

Melanoma
Background: PV-10 (rose bengal disodium) is the first small molecule oncolytic immunotherapy in development for solid tumors. Intralesional injection can yield immunogenic cell death and tumor-specific reactivity in circulating T-cells [1][2][3][4]. PV-10 is currently the subject of a Phase 1b/2 study in combination with systemic immune checkpoint inhibition (CI) for patients with advanced melanoma. Materials and methods: Study PV-10-MM-1201 (NCT02557321) is assessing PV-10 in combination with pembrolizumab. Patients must have at least 1 injectable lesion, at least 1 measurable target lesion (TL), and be candidates for pembrolizumab. In the Phase 1b portion of the study, patients receive combination treatment during the induction phase (q3w for 5 cycles) and then pembrolizumab alone in the maintenance phase (total duration of up to 24 months); the primary endpoint is safety and tolerability with objective response rate (ORR) and progression-free survival as key secondary endpoints (assessed via RECIST 1.1 after 15 weeks then q12w). Results: An initial Phase 1b cohort of predominantly CI-naïve subjects reached full accrual in April 2018 (20 Stage IV and 3 Stage IIIC/ IIID patients). All Treatment-Emergent Adverse Events (TEAEs) were consistent with established patterns for both drugs, with no significant overlap of AEs or unexpected toxicities. All disease stages exhibited response after minimal PV-10 intervention (median 4 cycles of PV-10, range 1-5; median 5 injections of PV-10 per patient, range 1-82), with 9% complete response (CR) and 65% ORR (overall per RECIST) as of a 1 Nov 2018 data cutoff. Response of injected target lesions (77% CR and 80% ORR across all disease stages) was higher than historical data for single-agent PV-10 (46% CR and 53% ORR across all disease stages). Although data on the combination for treatment of Stage III (M0) disease is currently limited, response rate of injected target lesions was also higher for M0 disease than observed in single-agent use: 67% CR (4 of 6 lesions) vs 54% CR (214 of 395 lesions). Conclusion: Robust response was observed across all disease stages. A first expansion cohort is accruing CI-refractory patients to further characterize response in this emergent population. Systemic therapy with CI is now recommended in the USA for Stage III patients with satellite or in-transit disease [5], but KEYNOTE-001 demonstrated lower overall response in M0 vs M1 patients [6] and for subcutaneous vs visceral lesions [7]. To address this population, a second expansion cohort directed to patients with satellite or in-transit disease will be opened in early-2019.