Abstract: Introduction: Nivolumab (nivo) is a fully human IgG4 monoclonal antibody (mAb) targeting programmed death receptor-1 (PD-1). Nivo has demonstrated clinical activity and an acceptable safety profile in a phase 1b study (NCT01592370; CheckMate 039) in patients (pts) with relapsed/refractory hematologic malignancies. In pts diagnosed with Hodgkin lymphoma (HL), after a median 86 weeks of follow-up, 7/20 responders maintained a response for 〉 1.5 years (Ansell S et al. Blood 2015;126:583), and after a median follow-up of 67 weeks, clinical activity (investigator-assessed objective response rate) was demonstrated in follicular lymphoma (FL; 40%), diffuse large B-cell lymphoma (DLBCL; 36%), mycosis fungoides (15%), and peripheral T-cell lymphoma (PTCL; 40%) (Lesokhin AM et al. J Clin Oncol 2016;34:2698). CheckMate 039 also included a cohort of pts who had received nivo in combination with ipilimumab (ipi), a fully human mAb targeting cytotoxic T-lymphocyte antigen 4 (CTLA-4). Combination of CTLA-4 and PD-1 blockade has shown superior efficacy compared with nivo or ipi alone in preclinical studies and solid tumor malignancies (Wolchok JD et al. N Engl J Med 2013;369:122; Larkin JM et al. N Engl J Med 2015;373:22; Antonia SJ et al. Lancet Oncol 2016;17:883). The aim of this cohort study was to evaluate the safety and efficacy of combined immune checkpoint blockade (nivo+ipi) in pts with the following hematologic malignancies: HL, B-cell non-Hodgkin lymphoma (B-NHL; FL and DLBCL), T-cell NHL (T-NHL; cutaneous T-cell lymphoma [CTCL] and PTCL] ), and multiple myeloma (MM). Methods: Nivo+ipi were given at 3 mg/kg IV and 1 mg/kg IV, respectively, every 3 weeks for 4 doses, followed by nivo monotherapy (3 mg/kg) every 2 weeks for up to 2 years. Pts with any of the above histologies, relapsed or refractory disease after ≥2 prior lines of therapy, and adequate organ function were included in the study. Prior systemic therapy may have included chemotherapy and autologous hematopoietic stem cell transplantation (auto-HSCT). Prior anti-PD-1 therapy and allogeneic (allo)-HSCT were not permitted. The primary endpoint was safety. Secondary endpoints included investigator-assessed objective response rate (ORR), best overall response, duration of response (DOR), and progression-free survival (PFS). Results: In total, 65 pts were treated with nivo+ipi (31 HL, 15 B-NHL, 11 T-NHL, 7 MM, and 1 with primary mediastinal B-cell lymphoma [PMBL] who was included in the overall safety cohort only). Median (range) number of prior systemic therapies was 4 (2, 10; HL), 3, (1, 16; B-NHL), 4 (1, 11; T-NHL), and 5 (2, 20; MM). Among patients with HL, only 13% (4/31) had prior auto-HSCT. 2 pts with HL and 1 with T-NHL proceeded to allo-HSCT after stopping study therapy. Across all cohorts, median follow-up was 11.4 months. 5 pts (8%) discontinued due to a drug-related adverse event (AE). The most common drug-related AEs of any grade were fatigue (17 pts [26%] ), pyrexia (15 [23%]), and diarrhea (12 [18%] ). 19 pts (29%) had a drug-related AE of grade ≥3. 31 pts (48%) had a serious AE. 24 pts (37%) died: HL 2 pts, B-NHL 11, T-NHL 6, MM 4, PMBL 1. Among those pts, 22 (34%) were from disease progression (HL 2 pts, B-NHL 10, T-NHL 5, MM 4, PMBL 1); no deaths were due to an AE. Clinical outcome data are presented (Table). Conclusions: These are the first reported data of combination checkpoint blockade therapy in hematologic malignancies. Overall, the combination of nivo+ipi in these heavily pretreated patients demonstrated a safety and efficacy profile similar to that previously reported for nivo monotherapy in HL, NHL, and MM. Additional follow-up may further clarify the role of ipi in this cohort of patients. In this predominantly transplant-naïve group of patients with HL, the efficacy of nivo+ipi was similar to that seen in patients with relapsed/refractory HL treated with nivo alone. Funding: Bristol-Myers Squibb (BMS). Medical writing: S Addison, Caudex, funded by BMS Disclosures Ansell: BMS, Seattle Genetics, Merck, Celldex and Affimed: Research Funding. Gutierrez:Bayer Health Care Pharmaceuticals, Inc.: Other: Traveling and Lodging- Food and Beverage; E.R. Squibb & Sons, LLC (Bristol Myers Squibb): Consultancy, Other: Travel and Lodging; Incyte Corporation: Consultancy; Pfizer Inc: Consultancy; Merck Sharp & Dohme Corporation: Consultancy, Other: Travel and Lodging; Pharmacyclics LLC, An AbbVie Company: Other: Food and Beverage. Shipp:Cell Signaling: Honoraria; Bristol-Myers Squibb: Consultancy, Research Funding; Bayer: Research Funding; Merck, Gilead, Takeda: Other: Scientific Advisory Board. Moskowitz:Seattle Genetics: Research Funding; Seattle Genetics, Merck: Consultancy. Borello:Bristol-Myers Squibb: Research Funding, Speakers Bureau. Popa-Mckiver:Bristol-Myers Squibb: Employment, Equity Ownership. Farsaci:Bristol-Myers Squibb: Employment. Zhu:Bristol-Myers Squibb: Employment. Armand:Sequenta Inc: Research Funding; Pfizer: Research Funding; Bristol-Myers Squibb: Consultancy, Research Funding; Infinity Pharmaceuticals: Consultancy; Roche: Research Funding; Merck: Consultancy, Research Funding.

Abstract: We have prospectively collected data on Adverse Events (AE) that occurred in 179 Hemopoietic Progenitor Cell (HPC) infusions performed in patients affected with haematological neoplasm, after high dose chemotherapy. Stem cell source was Hemopoietic Progenitor Cells Aphaeresis (HPC-A) in 157 cases and Hemopoietic Progenitor Cells Bone Marrow (HPC-BM) in 22 cases. In all cases, an endotoxin-free DMSO was used. One or more AE were registered in 51/179 infusions (28.6%). Frequency of AE was higher after HPC-A than after HPC-BM (31.3% versus 4.5%, (chi square test: p=0.008). In univariate logistic regression other factors found important for AE were: Age (p=0.028), Number of Total Nucleated Cells infused/kg (P=0.002), Volume/kg infused (p=0.057), Volume of Packed Red Blood Cells (p=0.019), a content of Non-Mononuclear Cells 〉 0.500 × 108/Kg ( 〈 p=0.0001) and Actual Time of infusion (p=0.058). When all aforementioned factors were evaluated in multivariate logistic regression only Age of patient (P=0.024) and a content of Non-Mononuclear Cells 〉 0.5×108/kg (P=0.0003) remained significant. No cardiovascular events were recorded during infusions. A significant correlation existed between reduction of cardiac frequency both with Volume/Kg infused (r 0.221; p=0.02) and with Actual Time of infusion (r 0.269; p=0.005). In conclusion, while Cardiovascular Changes are influenced by Volume/Kg infused and by Actual Time of infusion, Non-Cardiovascular AE are dependent on patient Age and on contamination by Non-Mononuclear Cells in apheretic harvests.

Abstract: Background: Several monoclonal antibodies (MAbs) with demonstrated clinical anti-cancer activities have been engineered as fully human IgG1 entities to also encompass their potential to mediate antibody-dependent cell-mediated cytotoxicity (ADCC) of human tumor cells. MSB0010718C is a fully human IgG1 MAb targeting the co-regulatory protein Programmed Death-Ligand 1 (PD-L1), and is thus distinct from other MAbs targeting the PD-L1/PD-1 axis currently being evaluated in clinical trials. One possibility is that an anti-PD-L1 antibody capable of inducing ADCC may negatively affect PD-L1 expressing immune cell subtypes. This work is intended to determine if there is any validity to this concern. Methods: The clinical activity of MSB0010718C, observed in several tumor types in ongoing clinical studies such as NCT01772004, has been and will be reported elsewhere. In the studies reported here, MSB0010718C is shown to mediate ADCC of several types of human tumor cell lines (e.g., breast, lung, bladder carcinomas) in vitro, with tumor cell lysis mediated mainly by human CD16+ monocytes and natural killer (NK) cells. Since some human immune cell subsets express PD-L1 on their cell surface (albeit at relatively low levels compared to many tumor cells), studies were undertaken to evaluate changes in the frequency of immune cell subsets in peripheral blood mononuclear cells (PBMC) from cancer patients pre- vs post-treatment with MSB0010718C. Immune cells evaluated were PD-L1 positive and PD-L1 negative subsets of the following: CD4+ T cells, CD8+ T cells, NK cells, regulatory T cells (Tregs), myeloid-derived suppressor cells (MDSC), natural killer T cells (NKT), plasmacytoid dendritic cells (DC), conventional DC, and B cells. Results: Forty-two post-treatment PBMC samples were evaluated as follows: pre vs 1 dose of MSB0010718C (day 15, n = 19); pre vs 3 doses of MSB0010718C (day 43, n = 14); and pre vs 9 doses of MSB0010718C (day 127, n = 9). In all cases there were no statistical differences pre- vs post-treatment in any immune cell subset, and at any time point analyzed, regardless of whether the immune subset expressed PD-L1 or not. In addition, no changes were observed in absolute lymphocyte counts at any time point analyzed. Conclusion: While immune cell subsets pre- vs post-treatment continue to be analyzed in various patient cohorts, these studies provide evidence that MSB0010718C, a fully human IgG1 MAb, capable of mediating ADCC, can be administered safely to cancer patients without altering the balance of numerous PBMC immune cell subsets. Citation Format: Lauren M. Lepone, Renee N. Donahue, Benedetto Farsaci, Italia Grenga, Benjamin Boyerinas, Caroline Jochems, Kwong-Yok Tsang, Christopher R. Heery, Ravi A. Madan, Geraldine O'Sullivan Coyne, Harpreet Singh, James L. Gulley, Jeffrey Schlom. Evaluation of immune cell subsets of cancer patients treated with a fully human IgG1 anti-PD-L1 MAb (MSB0010718C) capable of mediating ADCC of human tumor cells. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1316. doi:10.1158/1538-7445.AM2015-1316

Abstract: Many tyrosine kinase inhibitors (TKIs) that have antiangiogenic proprieties have also shown to have immunomodulatory proprieties by decreasing the number and function of immunosuppressive elements such as myeloid-derived suppressor cells and T-regulatory lymphocytes. These immunomodulatory properties make TKIs attractive candidates for combination with cancer immunotherapy. The aim of this study was to investigate the differential action of vaccine and the antiangiogenic TKI sunitinib on tumor architecture, to better define the mechanisms underlying the clinical benefit of this combination. Using a CEA+ colon carcinoma mouse model in CEA-transgenic mice, we show for the first time that vaccine and TKI exert differential effects on tumor architecture. Fluorescent immunohistochemistry (IHC) of 3 distinct types of tumor vessels (identified as CD31+, CD105+, and the monocytic marker CD11b+) demonstrated that vaccine decreased monocytic vasculature while sunitinib decreased the 3 types of vasculature and fenestrations. Immunoenzymatic IHC indicated that each single treatment and the combination affected tumor compactness and tight junctions. Each single treatment decreased hypoxia, as shown by hypoxyprobe IHC analysis, while combining sunitinib with vaccine resulted in the highest tumor oxygenation. Finally, vaccine plus sunitinib increased the tumor infiltration of antigen-specific T lymphocytes. These observations suggest that vaccine acted not only as an immune stimulant, but also as an angiogenic agent. These insights will assist in rationally developing and implementing such combination therapies. Citation Format: Renee N. Donahue, Italia Grenga, Peter S. Kim, Michael A. Coplin, James W. Hodge, Jeffrey Schlom, Benedetto P. Farsaci. Immunomodulatory effects of a tyrosine kinase inhibitor and vascular remodeling properties of a cancer vaccine potentiate combinatorial immunotherapy. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1232. doi:10.1158/1538-7445.AM2013-1232

Abstract: Classical Hodgkin lymphomas (cHLs) include infrequent malignant Hodgkin Reed-Sternberg (HRS) cells within an extensive but ineffective inflammatory/immune cell infiltrate. HRS cells exhibit frequent copy number alterations (CNAs) of 9p24.1/CD274 (PD-L1)/PDCD1LG2(PD-L2), ranging from low-level polysomy to relative copy gain and high-level amplification, all associated with increased expression of programmed death receptor-1 (PD-1) ligands on the tumor cell. PD-1 ligands engage the PD-1 receptor on T cells, inhibiting T cell activation and anti-tumor immune responses. cHL patients (pts) with the highest-level 9p24.1 alterations, PD-L1/PD-L2 amplification, have inferior progression-free survival (PFS) after standard primary chemotherapy (Roemer et al. J Clin Oncol 2016). Given the demonstrated responsiveness of cHL to PD-1 blockade, we examined the prevalence and type of 9p24.1 genetic alterations, PD-L1 expression, and association of these alterations with clinical outcome in pts receiving nivolumab (nivo; anti-PD-1) for relapsed/refractory (R/R) cHL. CheckMate205 is a multicenter, multicohort, phase 2 trial of nivo in R/R cHL. This analysis focused on 2 cohorts: pts with recurrent cHL following autologous stem cell transplantation (ASCT) and subsequent brentuximab vedotin (BV) (cohort B), and pts with R/R cHL following ASCT and BV given pre- or post-ASCT (cohort C). Pts received nivo 3 mg/kg every 2 weeks. Best overall response (BOR) and PFS were assessed by an independent radiological review committee (IRRC). In pts with available tumor biopsies, 9p24.1 genetic alterations were evaluated via fluorescence in situ hybridization (FISH) assay; probes encompassed CD274 (PD-L1, red) or PDCD1LG2 (PD-L2, green)and included a centromeric control (aqua).Dual immunohistochemical staining of PD-L1/PAX5 was performed to delineate PD-L1 expression in PAX5dim+ HRS cells and PAX5- cells in the tumor microenvironment. A modified PD-L1 H-score (range 0-300) was calculated by multiplying the percentage of PAX5+ (malignant) or PAX5- (non-malignant) cells with positive staining (0-100%) and the average intensity of positive staining (1-3+; ≥50 RS cells counted). All p-values are nominal. 96 pts had evaluable baseline tumor biopsy specimens; all 96 had detectable 9p24.1 alterations: polysomy in 10/96 (10%), copy gain in 56/96 (58%), amplification in 28/96 (29%), and presumptive rearrangement (split-apart FISH signal) in 2/96 (2%) (Figure). There was a significant association between PD-L1 protein expression (H-score) and the level of 9p24.1 alterations in HRS cells (p=0.002) (Figure). We next evaluated the association between BOR and defined 9p24.1 alterations and PD-L1 H-scores. The level of 9p24.1 CNAs was significantly associated with BOR (p=0.01); no pts with progressive disease (PD) had genomic amplification and no pts with complete response (CR) had polysomy. Similarly, there was significant association between PD-L1 H-score in HRS cells and BOR (p=0.02); all pts with PD had PD-L1 H-scores in quartiles (Q) 1/2, whereas most pts with CR had PD-L1 H-scores in Q4. We also assessed the association of PFS with 9p24.1 status and PD-L1 H-scores. The level of 9p24.1 alterations was associated with PFS; pts with 9p24.1 amplification had the most favorable PFS with nivo. HRS cell PD-L1 H-score (by quartiles) was also significantly associated with PFS (p=0.048), as pts with PD-L1 H-scores in Q1/Q2 had significantly shorter PFS than those with scores in Q3/Q4 (p=0.01). In contrast, there was no association between PD-L1 expression on non-malignant cells in the tumor microenvironment and PFS (p=0.43). In conclusion, all evaluable pts in this study had genetic alterations of 9p24.1/PD-L1/PD-L2 and copy number-dependent increased expression of PD-L1 in HRS cells. Although high-level alterations of 9p24.1 and increased PD-L1 expression were previously linked with inferior response to standard primary chemotherapy, we now associate these parameters with more favorable outcomes to targeted PD-1 blockade. These analyses also highlight the importance of quantifying and delineating PD-L1 expression in HRS cells and non-malignant cells in the tumor microenvironment in cHL biopsy evaluation. While further research is needed to guide possible use in clinical practice, these data advance understanding of 9p24.1 alterations and PD-L1 expression as prognostic biomarkers in cHL. Disclosures Engert: Takeda, BMS: Consultancy, Honoraria, Research Funding. Zinzani:Abbvie: Membership on an entity's Board of Directors or advisory committees; Janssen: Membership on an entity's Board of Directors or advisory committees; MorphoSys: Membership on an entity's Board of Directors or advisory committees; Takeda: Membership on an entity's Board of Directors or advisory committees; Celegene: Membership on an entity's Board of Directors or advisory committees; Roche: Membership on an entity's Board of Directors or advisory committees. Timmerman:Bristol-Myers Squibb, Kite Pharma, Valor Biopharmaceuticals, Janssen: Research Funding; Seattle Genetics, Genmab, Celgene: Consultancy, Honoraria. Ansell:BMS, Seattle Genetics, Merck, Celldex and Affimed: Research Funding. Armand:Bristol-Myers Squibb: Consultancy, Research Funding; Pfizer: Research Funding; Sequenta Inc: Research Funding; Roche: Research Funding; Merck: Consultancy, Research Funding; Infinity Pharmaceuticals: Consultancy. Kuruvilla:BMS: Consultancy, Honoraria; Abbvie: Consultancy, Honoraria; Gilead: Consultancy, Honoraria; Amgen: Honoraria; Celgene: Consultancy, Honoraria; Merck: Honoraria; Roche Canada: Consultancy, Honoraria, Research Funding; Janssen: Consultancy, Honoraria; Seattle Genetics: Consultancy, Honoraria; Lundbeck: Honoraria. Cohen:Bristol-Myers Squibb: Research Funding; Celgene: Consultancy, Membership on an entity's Board of Directors or advisory committees; Seattle Genetics: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Millennium/Takeda: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding; Pharmacyclics: Consultancy, Membership on an entity's Board of Directors or advisory committees; Infinity: Consultancy, Membership on an entity's Board of Directors or advisory committees; Novartis: Consultancy, Membership on an entity's Board of Directors or advisory committees, Research Funding. Collins:Takeda: Consultancy, Honoraria, Speakers Bureau. Trneny:Roche, Celgene, Takeda, Janssen, Gilead, Bristol-Myers Squibb: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Roche, Celgene: Research Funding. Farsaci:Bristol-Myers Squibb: Employment. Kato:Bristol-Myers Squibb: Employment. Sumbul:Bristol-Myers Squibb: Employment. Rodig:Bristol-Myers Squibb: Honoraria, Research Funding; Perkin Elmer: Membership on an entity's Board of Directors or advisory committees. Shipp:Bayer: Research Funding; Merck, Gilead, Takeda: Other: Scientific Advisory Board; Cell Signaling: Honoraria; Bristol-Myers Squibb: Consultancy, Research Funding.

Abstract: Background: Low CD34+ cell mobilization in P.B. has been found in a quota of AML patients (10–30%). Contrary to what has been observed in CD34+ mobilization in other haematological afflictions, in AML no features pertaining to the disease or to the patient have been found to be predictive of CD34+ cell mobilization failure. A possible explanation for this particular aspect of CD34+ mobilization in AML patients could be an intrinsic abnormality of non leukemic hematopoietic cells determining an increased chemo-sensitivity to anti-neoplastic drugs. To test this hypothesis we assessed, in AML patients, frequency of various types of clonable precursors (CFUs) present in BM at the time of CR and their in vitro chemosensitivity. We also correlated this data with the efficiency of CD34+ cell mobilization in P.B. Methods: 31 consecutive patients, affected with AML and a group of 15 normal BM donors were prospectively studied. Baseline CFU-GEMM, BFU-E, CFU-GM, CFU-E as well as the sensitivity of these precursors to two chemotherapeutic agents (ASTA-Z and VP-16) were assayed on BM cells obtained in first CR after consolidation chemotherapy. Chemo-sensitivity (100 - normalized residual CFU) was studied after short term in vitro incubation of bone marrow precursors at various drug concentrations. All pts underwent a CD34+ mobilization attempt and, as measure of mobilization strength, peaks of CD34+ cells reached in P.B. were determined. Results: In AML patients, after induction and consolidation schedules, a reduced number of all types of CFUs were found in BM compared to normal controls. The frequency of any types of CFUs and the chemo-sensitivity of CFU-GEMM, BFU-E and CFU-E were not correlated to CD34+ peak reached in P.B. However, in AML patients, an inverse correlation was found between chemo-sensitivity of CFU-GM and maximum CD34+ cells peak reached in P.B. during mobilization (r= − 0.807 and p=0.0001, when ASTA-Z was used at 100 mcg/ml). In univariate and multivariate logistic regression, chemo-sensitivity to ASTA-Z of CFU-GM was the only factor significantly associated with mobilization failure (P=0.02), independently of age and cytogenetical risk. Chemosensitivity of CFU-GEMM, BFU-E and CFU-E after in vitro incubation with chemotherapeutic drugs was not different in AML patients compared to CFU obtained from normal control. The contrary was found for CFU-GM and, overall, CFU-GM from AML patients had a significantly higher chemosensitivity to ASTA-Z compared to CFU-GM of normal controls (p= 0.01 at 50 mcg/ml). Conclusions: We found that an abnormal high chemo-sensitivity of CFU-GM to some chemotherapy drugs in AML patients is associated with a high risk of CD34+ cell mobilization failure. This abnormality of non leukaemic bone marrow cells, present in CR, is restricted only to CFU-GM and is not evident in other CFUs. Figure Figure

Abstract: TPS8555 Background: Adoptive cellular therapy may be practice-changing in relapsed/refractory multiple myeloma (MM). NY-ESO-1 TCR T (GSK3377794) are autologous polyclonal T cells transduced by a self-inactivating lentiviral vector to express an affinity-enhanced TCR capable of recognizing NY-ESO-1 or LAGE-1a antigenic peptides in complex with HLA-A*02. GSK3377794 has shown clinical activity in synovial sarcoma, melanoma, myxoid/round cell liposarcoma, and MM after autologous stem cell transplant. NY-ESO-1 and LAGE-1a are cancer/testis antigens frequently overexpressed in MM and linked to poor clinical outcome. PD-1 expression on CD8 T cells, which has been observed in MM patients previously treated with GSK3377794 as well as with CD19 CAR T-cell therapy, can limit adaptive immune response. We hypothesize that GSK3377794 alone, or in combination with the anti-PD-1 inhibitor pembrolizumab, may result in an antitumor effect in MM. Methods: This is an open-label, pilot study (NCT03168438) of GSK3377794 in patients with relapsed/refractory MM positive for HLA-A*02:01, HLA-A*02:05, ± HLA-A*02:06 and NY-ESO-1/LAGE-1a. Patients (n = 20) who have received ≥3 prior therapies containing ≥1 immunomodulatory imide, proteasome inhibitor, alkylator, CD38 monoclonal antibody, or glucocorticoid will be assigned to either single-infusion GSK3377794 (Arm 1, n = 10) or single-infusion GSK3377794 + pembrolizumab 200 mg IV every 3 weeks (Arm 2, n = 10). Arm 1 enrollment will be completed first. In Arm 2, pembrolizumab will begin in Week 3 (Week 6 if precluded by toxicity). Patients in both arms will provide cells via leukapheresis to manufacture autologous NY-ESO-1–specific T cells, undergo lymphodepletion (fludarabine + cyclophosphamide), and then receive GSK3377794 infusion (1−8x10 9 transduced T cells). Primary and secondary objectives are to assess safety/tolerability and antitumor activity, respectively, of GSK3377794 (± pembrolizumab). Arm 2 enrollment will pause for a 3-week safety review after 3 patients have received a first dose of pembrolizumab. Efficacy, safety, and biomarkers will be assessed every visit. The treatment phase will last 108 weeks, or until disease progression; follow-up will last ≤15 years. As of January 2020, 3 patients have been treated. Funding: GlaxoSmithKline (208470) Clinical trial information: NCT03168438 .