Background

CT26 tumors are heavily infiltrated by tumor-specific CD8 T cells but nevertheless grow rapidly in Balb/c mice. These tumors are partially responsive to anti-PD-1 monoclonal antibody therapy, suggesting an active suppression of the tumor-specific cytotoxic T cells. As CCR2 is expressed by a potentially suppressive leukocyte subset within these these tumors, we aimed to test whether CCR2 blockade could enhance the anti-tumor effects of anti-PD-1.

Methods

Five days after subcutaneous CT26 implantation into the flanks of 9wk female Balb/c mice (2.5x10⁵/mouse), the recipients were randomized based on tumor size and treatment was begun. Mice received anti-PD-1 by IP injection on days 7, 10, 17 and 21 (200μg/mouse), and received CCR2 antagonist CCX598 (30 or 60mg/kg) or vehicle by oral gavage every 24 hours until day 60.

Results

We have found that the therapeutic effects of anti-PD-1 therapy are appreciably enhanced by specific blockade of chemokine receptor 2 (CCR2) via a small molecule antagonist. This combined anti-PD-1/CCR2i approach significantly decreases tumor size and increases the proportion of long-term survivors, with more than 50% of the mice (up to 73%) showing complete regression of a previously established tumor. The effects of this combined therapy are dependent on the presence of CD8⁺ T cells, as tumors do not respond to the therapy in CD8-depleted mice. The anti-CT26 tumor response is specific: long term survivors are resistant to re-inoculation with the CT26 tumor (even without further dosing of either drug) but are not resistant to the 4T1 breast tumor. CCR2 antagonism alters the tumor microenvironment by reducing the number of mMDSC per gram of tumor (a CCR2^hi population phenotypically defined as CD11b⁺/Ly6G^-/Ly6C^hi). Reduction in tumor size is inversely proportional to the ratio of CD8 T cells to mMDSC.

Conclusions

These data are consistent with a hypothesis that CCR2 antagonism enhances anti-PD-1 therapy by preventing mMDSC from accumulating within the tumor, thus reducing their suppressive effects on cytotoxic T cells.

Legal entity responsible for the study

ChemoCentryx, Inc.

Funding

ChemoCentryx, Inc.

Disclosure

J. Campbell, C. Janson, L. Ertl, C. Li, Z. Miao, V. Chhina, M. Vilalta, A. Kumamoto, T. Dang, S. Liu, S. Yao, P. Zhang, T.J. Schall, R. Singh: Full time employee of ChemoCentryx, Inc.

Background

The checkpoint inhibitor Ipilimumab induces long-lasting clinical responses in a proportion of patients with metastatic melanoma. However, most patients do not benefit from therapy. A desmoplastic stroma (tumour fibrosis), characterized by excessive extracellular matrix (ECM) remodeling, can result in reduced drug delivery and T-cell infiltration into the tumour and has been associated with poor prognosis and lack of therapy response. Here we address non-invasive biomarkers reflecting ECM remodeling in the tumour microenvironment, for associations with clinical response to Ipilimumab in patients with metastatic melanoma.

Methods

Serum biomarkers of collagen type III formation (Pro-C3), MMP-degraded type I collagen (C1M) and MMP-degraded type IV collagen (C4M) were studied with competitive ELISAs in pretreatment serum from 66 patients with metastatic melanoma. Results were correlated with objective response rate (ORR) and survival outcomes (OS) by univariate Cox analysis and Kaplan Meier plots.

Results

Pro-C3, C1M and C4M were significantly elevated in patients with progressive disease (PD) compared to the combined group of patients with stable disease (SD), partial response (PR) and complete response (CR) at baseline. Moreover, in the subgroup of patients with baseline biomarker levels higher than the 75^th percentile (Q4), only 6-13% responded (CR+PR+SD) compared to 46-48% in the group of patients with low biomarker levels (Q1-Q3). The median OS was 285, 161 or 198 days in biomarker high patients versus 596, 592 or 621 days in biomarker low patients for Pro-C3 (HR 2.13, 95%CI 1.12-4.04), C1M (HR 1.70, 95%CI 0.85-3.38) and C4M (HR 2.43, 95%CI 1.26-4.70), respectively.

Conclusions

High baseline levels of Pro-C3, C1M and C4M were associated with progressive disease and decreased overall survival in metastatic melanoma patients receiving Ipilimumab. This suggest the ECM biomarkers ability to predict response to checkpoint inhibitors at baseline. Future studies are needed to investigate the biomarkers applicability in other types of immunotherapies and cancers.

Legal entity responsible for the study

Nordic Bioscience

Funding

None

Disclosure

C. Jensen, M.A. Karsdal, N. Willumsen: Employee of Nordic Bioscience involved in biomarker development.

All other authors have declared no conflicts of interest.

Background

Despite originally considered an immunologically silent malignancy, recent studies are encouraging research of breast cancer immunogenicity to evaluate the applicability of immunotherapy as a treatment strategy. The epitope landscape in breast cancer is minimally described. Consequently, this project investigates four proteins commonly upregulated in breast cancer and thus probable tumor associated antigens (TAAs). Aromatase, prolactin, never in mitosis a related kinase 3 (NEK3), and protein inhibitor of activated STAT3 (PIAS3) contribute to increased growth, survival, and motility of malignant cells.

Methods

Aspiring to uncover novel epitopes for cytotoxic T cells, a reverse immunology approach was applied. In silico screening via NetMHC was used to predict binding of peptides within the proteins to HLA-A*0201 and -B*0702. Next, an MHC ELISA was applied to experimentally confirm which of the peptides are indeed HLA-A*0201 and -B*0702 binders. Hereafter, a novel method for high throughout detection of antigen specific T cells was applied. Via DNA barcode labeled MHC multimer technology, parallel screening for T cell recognition of all predicted MHC binding peptides was performed.

Results

415 peptides were predicted in silico as HLA-A*0201 and -B*0702 binders. Subsequent in vitro binding analysis confirmed binding for 147 of the 415 predicted binders. The 147 peptides were evaluated for T cell recognition utilizing DNA barcode labeled MHC multimers to screen peripheral blood lymphocytes from breast cancer patients and healthy donor samples. Significantly more TAA specific T cell responses were detected in breast cancer patients compared to healthy donors for both HLA-A*0201 (p < 0.0039) and -B*0702 (p< .A-0ealthy donors for both HLA-for both HLA A*0201 and B*0702 ses were detected in breast cancer patients 0.001) restricted peptides. Importantly, several of the identified responses were directed towards peptides that were predicted as poor or intermediate affinity binders. This is indicative of the importance of inclusion of low-affinity binders in the search for epitopes within shared TAAs, as these might be less subject to immune tolerance mechanisms.

Conclusions

Thus, the inspected proteins; aromatase, prolactin, NEK3 and PIAS3, indeed contain targets for T cell reactivity.

Legal entity responsible for the study

Sine Reker Hadrup

Funding

Danish Cancer Society, Danish Council for Independent Research

Disclosure

All authors have declared no conflicts of interest.

Background

Adoptive T-cell therapy (ACT) with tumor infiltrating lymphocytes (TIL) has proven to be a powerful treatment option for patients with metastatic melanoma with response rates of approximately 50% and durable complete responses in about 15%. However, there is still a need for improving TIL efficacy and a promising strategy is combination with immunomodulating agents. One such is vemurafenib (vem), a selective BRAF inhibitor, which induces objective responses in about 50% of melanoma patients with tumors expressing BRAF^V600E/K. In addition to the direct anti-cancer effect, vem has been shown to increase T-cell infiltration into tumors, upregulate melanoma antigen expression and increase the frequency of TIL recognizing autologous melanoma cells. This trial was previously presented at the Society for Immunotherapy of Cancer annual meeting (1,2). Updated data on clinical responses and preliminary immunological analyses will be presented.

Trial design

A total of 12 patients will be included in this open phase II non-randomized trial primarily to investigate safety when combining ACT and vem (ClinicalTrials.gov ID NCT02354690). Secondarily, clinical responses will be evaluated according to RECIST and extensive immune monitoring will be performed. Patients are treated with vem orally 960 mg BID one week prior to excision of tumor material for T-cell generation and continue vem until hospital admission (4-7 weeks). During hospitalization patients will receive a preparative lymphodepleting regimen consisting of cyclophosphamide 60 mg/kg for 2 days and fludarabine 25 mg/m² for 5 days. TIL infusion consists of 5-10 x 10¹⁰ T-cells and patients are subsequently treated with continuous interleukin-2 infusion following the decrescendo-regimen for 5 days. Patients are evaluated for up to 5 years or until progression. 1. Borch TH, Andersen R et al. T cell therapy in combination with Vemurafenib in BRAF mutated metastatic melanoma patients. J Immunother Cancer. 2014;2 (Suppl 3:P67). 2. Borch TH, Andersen R et al. T cell therapy in combination with Vemurafenib in BRAF mutated metastatic melanoma patients. Journal for ImmunoTherapy of Cancer 2016, 4(Suppl 1):P131.

Clinical trial identification

NCT02354690

Legal entity responsible for the study

Inge Marie Svane

Funding

Danish Cancer Society The Research Foundation of the Capital Region of Denmark

Disclosure

T.H. Borch: Honorarium for lecture from BMS. M. Donia: Honorarium from Genzyme, MSD, BMS. Received travel suppoer from Novartis, MSD, BMS, Roche and Pfizer. I.M. Svane: Advisory board memberships and lectures: Roche, Novartis, MSD, Celgene, Incyte, TILT bio, Pfizer, BMS, AstraZeneca. Institution has received limited grants for translational research from BMS, Roche, Novartis. All other authors have declared no conflicts of interest.

Background

E7046 is a selective small molecule antagonist of the prostaglandin E₂ receptor-type-4 that inhibits the differentiation of monocytic myeloid lineage cells towards a pro-tumorigenic phenotype in the TME. This is a first-in-human study of single agent E7046.

Methods

Key eligibility criteria: patients (pts) ≥18 years with selected advanced cancers with high levels of myeloid infiltrate. The dose-escalation phase consisted of 6-pt cohorts of 125, 250, 500, and 750 mg (once-daily, oral, 21-day cycle) doses of E7046. Primary objectives were safety/tolerability, maximum tolerated dose (MTD) and/or RP2D. Secondary objectives included PK and initial anti-tumor activity; exploratory objectives included PD assessments on immune cells in tumor infiltrate and in peripheral blood and metabolic response by ¹⁸FDG-PET.

Results

30 pts received E7046 (median age 58 yrs [24-78]; 2-7 lines of prior therapy). Most common tumor types were colorectal cancer (40%), pancreatic cancer (20%), and SCCHN (13%). No DLTs were observed and the MTD was not reached. The most frequent drug-related adverse events (AEs) were diarrhea (20%), decreased appetite, fatigue and nausea (13% each). Drug-related AEs of Gr 3/4 occurred in 4 pts (diarrhea, anaphylactic reaction, hypersensitivity, hyperuricemia, rash, generalized rash). 2 pts had drug-related serious AEs (rash, allergic reaction, fever in 1 pt; hyperuricemia, acute renal failure [Gr 2] in 1 pt). 3 pts discontinued treatment due to AEs (bowel obstruction, allergic reaction, abdominal pain). There were no drug-related deaths. E7046 exposure was dose proportional up to 500 mg with no incremental increase in exposure at 750 mg. E7046 was extensively metabolized, elimination half-life was ∼12hr and accumulation on multiple dosing was ∼2-fold. 2 pts are ongoing and preliminary efficacy showed no objective responses, 4 pts with durable SD or clinically stable (>4 mo) and 4 pts with ¹⁸FDG-PET metabolic responses.

Conclusions

Single-agent E7046 was tolerated with no MTD reached in heavily pretreated pts with myeloid-rich tumors. PD analysis of immune cell modulation to help determine the RP2D will be presented at the meeting.

Clinical trial identification

NCT02540291

Legal entity responsible for the study

Eisai Inc.

Funding

Eisai Inc.

Disclosure

D. Hong: Research/Grant Funding: Bayer, Lilly, Genentech, LOXO, Pfizer, Amgen, Mirati, Ignyta, Merck, Daichi-Sanko, Eisai; Travel, accommodations, expenses: MiRNA, LOXO; Consulting or Advisory Role: Bayer, Baxter, Guidepoint Global; Other ownership interests: Oncoreseponse (founder). A. Parikh: Personal fees from Roche, outside the submitted work. G. Shapiro: Consulting: Pfizer, Lilly, G1 Therapeutics, Vertex, Roche; Research funding: Lilly. L. Reyderman, M. Ren, T. Binder, C.E. Ooi: Employee of Eisai Inc. S. Dayal: Employee of Eisai Ltd. Ö. Ataman: Former employee of Eisai Ltd. at time of study. A. Marabelle: Received clinical trial funding from Eisai; Received consulting fees from Eisai and Roche; Received funding for anti-CSF1R clinical trial from Roche. All other authors have declared no conflicts of interest.

Background

The IFN-ɣ+CD4 Th1 response plays a critical role in anticancer immunity and has been extensively studied from tumor microenvironment in many cancers. Evidence support that blood represents a crossroads site in which the cancer immunity also takes place. Here we aimed to characterize the natural history of a tumor-specific Th1 immunity in blood that can provide relevant information about cancer evolution.

Methods

Peripheral blood mononuclear cells were collected from 170 patients with non-small cell lung cancer (NSCLC) before any treatment. The presence of IFN-ɣ+ CD4 T cell response against telomerase (TERT) was used as surrogate marker of the antitumor Th1. The anti-TERT Th1 response was measured by ELISpot after stimulation of PBMC with promiscuous epitopes from TERT. Flow cytometry and multiplex beads assay were used to measure multiple immune cells and soluble factors in blood.

Results

59 pts (35%) exhibited a pre-existing anti-TERT Th1 immunity. The frequency of anti-TERT Th1 responders decreased from localized to metastatic NSCLC (44.8% vs 24%, p < 0.05). The magnitude of anti-TERT Th1 response influenced the overall survival (OS), especially in metastatic patients: median OS of 16 vs 7 months in high and low anti-TERT Th1 responders respectively (p = 0.023). Among the blood immune parameters measured, T cells expressing PD-1 and/or TIM-3 were found inversely correlated to this response, so that blocking these receptors restored the anti-TERT Th1 cycle. Interestingly, the levels of anti-TERT Th1 combined to PD-1+/TIM-3+ CD4 T cells predict distinct immune profile. Patients with anti-TERT Th1^hi/CD4-PD-1/TIM-3^hi showed better OS as compared to in anti-TERT Th1^lo/CD4-PD-1/TIM-3^hi group (median OS: not reached vs 12 months respectively, p = 0.006). Thus, a strong anti-TERT Th1 immunity may counteract the poor prognosis associated with high rate of PD-1/TIM-3+ CD4 T cells.

Conclusions

Our data demonstrate a striking link between the systemic anti-TERT Th1 response with tumor burden and clinical outcome. It also provides evidence that the association of exhausted T cells and anti-TERT Th1 immunity monitoring could be used as a relevant blood-based dynamic immune biomarker.

Legal entity responsible for the study

Olivier Adotévi

Funding

Assistance Publique des Hopitaux de Paris (AP-HP), FEDER-FUI, Ligue Nationale contre le Cancer

Disclosure

All authors have declared no conflicts of interest.