Purpose: Relapsed and/or refractory acute myeloid leukemia and high-risk myelodysplastic syndrome continue to have a poor prognosis with limited treatment options despite advancements in rational combination and targeted therapies. Belinostat (an HDAC inhibitor) and Pevonedistat (a NEDD8 inhibitor) have each been independently studied in hematologic malignancies and have tolerable safety profiles with limited single-agent activity. Preclinical studies in AML cell lines and primary AML cells show the combination to be highly synergistic, particularly in high-risk phenotypes such as p53 mutant and FLT-3-ITD positive cells. Here, we present the safety, pharmacokinetics and pharmacodynamics of belinostat and pevonedistat in a dose escalation Phase I study in AML and High-Risk MDS. Methods: Eighteen patients (16 with AML, 2 with MDS) were treated at 5 dose levels (belinostat 800-1000mg/m2, pevonedistat 20-50mg/m2). Safety and tolerability were assessed according to protocol defined dose limiting toxicities (DLTs). Correlative pharmacokinetic and pharmacodynamic analyses were performed. Results: No dose limiting toxicities were noted. Most Grade 3 or 4 toxicities were hematologic in nature. The best response was stable disease in four patients, and complete remission in one patient who qualified as an exceptional responder. Pharmakokinetic studies revealed no association between drug exposure and best response. Pharmacodynamic RT-PCR studies demonstrated post-treatment increases in several proteins, including quantitative increases in the oxidative stress protein NQO1, ferroptosis protein SLC7A11, and GSR, linked to glutathione metabolism and oxidative stress, as did the anti-oxidants SRXN1 and TXNRD1. Conclusions: Patterns of post-treatment changes in correlative pharmacodynamic parameters may suggest possible mechanistic changes in the DNA damage response, oxidative damage, and ferroptosis pathways. The combination of pevonedistat plus belinosat is safe in an adult relapsed and/or refractory AML/High-Risk MDS population with modest but notable activity in this heavily treated, high risk population. Our findings also raise the possibility that certain extremely poor prognosis AML patients may respond to a regimen combining two targeted agents that have little or no activity when administered individually. Trial Registration: ClinicalTrials.gov ID NCT03772925, first posted 12/12/2018; CTEP Identifier 10246

Abstract Background Sapanisertib is a potent ATP‐competitive, dual inhibitor of mTORC1/2. Ziv‐aflibercept is a recombinant fusion protein comprising human VEGF receptor extracellular domains fused to human immunoglobulin G1. HIF‐1α inhibition in combination with anti‐angiogenic therapy is a promising anti‐tumor strategy. This Phase 1 dose‐escalation/expansion study assessed safety/ tolerability of sapanisertib in combination with ziv‐aflibercept in advanced solid tumors. Methods Fifty‐five patients with heavily pre‐treated advanced metastatic solid tumors resistant or refractory to standard treatment received treatment on a range of dose levels. Results Fifty‐five patients were enrolled and treated across a range of dose levels. Forty were female (73%), median age was 62 (range: 21–79), and ECOG PS was 0 (9, 16%) or 1 (46, 84%). Most common tumor types included ovarian (8), colorectal (8), sarcoma (8), breast (3), cervical (4), and endometrial (4). Median number of prior lines of therapy was 4 (range 2–11). Sapanisertib 4 mg orally 3 days on and 4 days off plus 3 mg/kg ziv‐aflibercept IV every 2 weeks on a 28‐day cycle was defined as the maximum tolerated dose. Most frequent treatment‐related grade ≥2 adverse events included hypertension, fatigue, anorexia, hypertriglyceridemia, diarrhea, nausea, mucositis, and serum lipase increase. There were no grade 5 events. In patients with evaluable disease (n = 50), 37 patients (74%) achieved stable disease (SD) as best response, two patients (4%) achieved a confirmed partial response (PR); disease control rate (DCR) (CR + SD + PR) was 78%. Conclusion The combination of sapanisertib and ziv‐aflibercept was generally tolerable and demonstrated anti‐tumor activity in heavily pre‐treated patients with advanced malignancies.

Background This investigator-initiated phase II trial aimed to evaluate the efficacy of cabozantinib in combination with nivolumab and ipilimumab (CaboNivoIpi) in previously treated patients with radioactive iodine-refractory differentiated thyroid cancer.Methods Eligible patients with radioactive iodine-refractory differentiated thyroid cancer who progressed on 1 prior line of vascular endothelial growth factor receptor-targeted therapy received a 2-week run-in of cabozantinib monotherapy followed by CaboNivoIpi for 4 cycles (cycle length = 6 weeks), followed by cabozantinib plus nivolumab (cycle length = 4 weeks) until disease progression. The primary endpoint was objective response rate (ORR) within the first 6 months of treatment. A Simon optimal 2-stage design allowed for an interim analysis after accrual of 10 evaluable patients. At least 5 responses were needed to proceed to stage 2.Results Among 11 patients enrolled, the median age was 69 years. Prior vascular endothelial growth factor receptor-targeted therapies included lenvatinib, pazopanib, and sorafenib plus everolimus. Median follow-up was 7.9 months. Among 10 evaluable patients, ORR within the first 6 months of treatment was 10% (1 partial response). Median progression-free survival was 9 months (95% CI, 3.0-not reached) and median overall survival was 19.2 months (95% CI, 4.6-not reached). Grade 3/4 treatment-related adverse events (AEs) were noted in 55% (6/11) and grade 5 AEs in 18% (2/11) of patients. The most common treatment-related AE was hypertension. The study did not reach its prespecified efficacy threshold.Conclusion CaboNivoIpi had low ORRs and a high rate of grade >= 3 treatment-related AEs.Clinical Trial Registration NCT03914300

Combination of DNA-PK inhibitors with trabectedin. A, Top, Mean Bliss matrix scores (n = 546) were calculated from the concentration matrix [(5 concentrations of drug A) × (6 concentrations of drug B) = (30 combination concentrations)] of each combination tested (n = 21) in all complex tumor spheroid models (n = 26). Mean Bliss matrix scores from the combination of the DNA-damaging drug trabectedin with the DNA-PK inhibitor nedisertib are highlighted in blue, whereas all other combinations are shown in light gray. A, Bottom, A mean Bliss score plot (n = 104) calculated from the combination concentration matrix of trabectedin with nedisertib across all complex tumor spheroid models (n = 26). The data are colored as a heat map, where blue indicates synergy, yellow indicates additivity, and red indicates antagonism. B, Concentration–response curves (top, mean ± SD, n = 4 technical replicates) from combinations of trabectedin with nedisertib and corresponding mean Bliss score plots (bottom, n = 4 technical replicates) showing the scores from each combination's concentration matrix and colored as a heat map. Data are shown for complex tumor spheroids grown with the malignant cell lines (from left): ASPS-1 (alveolar soft part sarcoma), HS-SY-2 (synovial sarcoma), and NCI-H211 (SCLC). C, Top, Mean Bliss matrix scores are highlighted from the combination of trabectedin with the DNA-PK inhibitor VX-984 across all complex tumor spheroid models (blue, n = 26), whereas all other combinations are shown in light gray. C, Bottom, A mean Bliss score plot (n = 104) calculated from the combination concentration matrix of trabectedin with VX-984 across all complex tumor spheroid models (n = 26). D, Concentration–response curves (top, mean ± SD, n = 4 technical replicates) from combinations of trabectedin with VX-984 and corresponding mean Bliss score plots (bottom, n = 4 technical replicates) showing the scores from each combination's concentration matrix and colored as a heat map. Data are shown for complex tumor spheroids grown with the malignant cell lines (from left): ASPS-1 (alveolar soft part sarcoma), HS-SY-2 (synovial sarcoma), and NCI-H211 (SCLC).

Combination of AZD-1390 and topotecan or trabectedin. A, Mean Bliss matrix scores (n = 546) were calculated from the concentration matrix [(5 concentrations of drug A) × (6 concentrations of drug B) = (30 combination concentrations)] of each combination tested (n = 21) in all complex tumor spheroid models (n = 26). The mean Bliss matrix scores are highlighted from combinations of the ATM inhibitor AZD-1390 with DNA-damaging drugs topotecan or trabectedin (n = 52, orange), whereas the scores from all other combinations are shown in light gray. B, A scatter plot of the mean Bliss matrix scores from the AZD-1390 combinations with topotecan and trabectedin (Pearson r = 0.77, two-tailed P < 0.0001). Concentration–response curves (top, mean ± SD, n = 4 technical replicates) from combinations of AZD-1390 with either topotecan (C) or trabectedin (D) and corresponding mean Bliss score plots (bottom, n = 4 technical replicates) showing the scores from each combination's concentration matrix and colored as a heat map (blue indicates synergy; yellow indicates additivity; red indicates antagonism). Data are shown for complex tumor spheroids grown with the malignant cell lines (from left): G-401 (rhabdoid tumor), NCI-H841 (SCLC), 287954-098-R-J1 (Ewing sarcoma), ASPS-1 (alveolar soft part sarcoma), and DMS 114 (SCLC).

Figure S1. Representative brightfield images for assay optimized cell densities from a DMSO-treated well on Day 10.

Table S1. Top-ranking drug combinations by mean Bliss matrix score.

Background: Radiation is a potent inducer of DNA double-strand breaks, and ribonucleotide reductase (RNR) is the rate-limiting enzyme in the synthesis and repair of DNA, making RNR-targeted therapy a rationale therapeutic strategy for radiosensitization. ETCTN 10388 (NCT04234568) evaluated safety and efficacy of the combination of lutetium 177 DOTATATE, a beta-emitting radionuclide in combination with triapine, a ribonucleotide reductase (RNR) inhibitor. Method: This study was a multicenter phase 1 dose escalation trial [using the Bayesian optimal interval design (BOIN)] of triapine in combination with fixed dose lutetium Lu 177 DOTATATE for well-differentiated somatostatin receptor-positive gastroenteropancreatic neuroendocrine tumor (GEP-NETs) after the failure of at least one line of prior systemic cancer treatment with an expansion cohort at the recommended phase 2 dose (RP2D). Oral triapine (100mg, 150mg, 200mg) was administered once daily on days 1-14 and Lu-177 DOTATATE [200 mCi] intravenously on day 1 of every 56-day cycle. A total of 4 cycles were administered. Response and adverse effects were assessed per RECIST and CTCAE 5.0, respectively. Exploratory correlative studies included tumor somatic and germline mutation testing, RNA sequencing, pharmacokinetics, deoxynucleosides and circulating cell free DNA analysis. Primary endpoints were safety and RP2D. Results: Overall, 31 patients were enrolled between 6 sites, 15 in the dose escalation phase and 16 in the dose expansion phase. Adverse events (AE) were assessed in all 31 patients per CTCAE 5.0. One DLT in dose level 1, seven DLTs in dose level 2, and one grade 5 DLT in dose level 3 were observed. The RP2D of the combination is triapine 150 mg QD (dose level 2) on days 1-14 in combination with Lu-177 DOTATATE on day 1 of every 56-day cycle. Detailed safety and adverse event data will be presented at the meeting. There were 28 patients evaluable for efficacy, of which 6 (21%) achieved a partial response. At 12 months, 6 patients had progressed, while 22 (86%) remained progression free. Median PFS has not been reached. PK data were available for 12 patients enrolled in the dose escalation cohort. The geometric mean (SD) AUC0-inf was 1159 (1.22) µg/L•h for the 100mg dose level and 1862 (1.76) µg/L•h for the 150 mg dose level, suggesting that exposure increased with dose, and inter-patient variability was as expected for an oral agent. Conclusion: The combination of triapine and Lu-177 DOTATATE was safe with preliminary efficacy signals, which will be further evaluated in ETCTN 10558, a randomized phase 2 study that is comparing the effectiveness of triapine and Lu-177 DOTATATE to Lu-177 DOTATATE alone. Citation Format: Aman Chauhan, Susanne Arnold, Jill Kolesar, William Carson, Heidi Weiss, Rani Jayswal, Donglin Yan, Riham El Khouli, Aman Khurana, Jan Beumer, Heloisa Soares, Mary Mulcahy, Thorvardur Halfdanarson, Daneng Li, Heather Jacene, Percy Ivy, Elise Kohn, John Wright, Larry Rubinstein, Zeta Chow, Piotr Rychahou, Mark B. Evers, Charles Kunos, Lowell Anthony, Bhavana Konda. ETCTN 10388: a first in human phase I trial of triapine and lutetium Lu 177 DOTATATE in well-differentiated somatostatin receptor-positive gastroenteropancreatic neuroendocrine tumors (GEP-NETs) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr CT194.

The malignant cell lines grown as complex tumor spheroids for this study. The names of both patient-derived and established cell lines are listed along with the tumor type they were derived from and key genetic alterations

Table S2. Links to the PubChem assays.

Figure S4. Heat maps of Bliss synergy scores across the combination dose-response matrices for all twenty-six cell lines grown as multicellular complex spheroids exposed to each DNA damaging agent (A, TMZ; B, topotecan; C, trabectedin) in combination with ATR inhibitors, berzosertib or elimusertib.

Multicellular spheroids comprised of malignant cells, endothelial cells, and mesenchymal stem cells served as an in vitro model of human solid tumors to investigate the potentiation of DNA-damaging drugs by pharmacologic modulation of DNA repair pathways. The DNA-damaging drugs, topotecan, trabectedin, and temozolomide were combined with varied inhibitors of DNA damage response enzymes including PARP (olaparib or talazoparib), ATM (ataxia telangiectasia mutated; AZD-1390), ATR (ataxia telangiectasia and Rad3-related protein; berzosertib or elimusertib), and DNA-PK (DNA-dependent protein kinase; nedisertib or VX-984). A range of clinically achievable concentrations were tested up to the clinical Cmax, if known. Mechanistically, the types of DNA damage induced by temozolomide, topotecan, and trabectedin are distinct, which was apparent from the response of spheroids to combinations with various DNA repair inhibitors. Although most combinations resulted in additive cytotoxicity, synergistic activity was observed for temozolomide combined with PARP inhibitors as well as combinations of the ATM inhibitor AZD-1390 with either topotecan or trabectedin. These findings might provide guidance for the selection of anticancer agent combinations worthy of further investigation. Significance: Clinical efficacy of DNA-damaging anticancer drugs can be influenced by the DNA damage response in tumor cells. The potentiation of DNA-damaging drugs by pharmacologic modulation of DNA repair pathways was assessed in multicellular tumor spheroids. Although most combinations demonstrated additive cytotoxicity, synergistic cytotoxicity was observed for several drug combinations.

Figure S5. Heat maps of Bliss synergy scores across the combination dose-response matrices for all twenty-six cell lines grown as multicellular complex spheroids exposed to each DNA damaging agent (A, TMZ; B, topotecan; C, trabectedin) in combination with DNA-PK inhibitors, nedisertib or VX-984.

Combination of ATR inhibitors with topotecan or trabectedin. Concentration–response curves (top, mean ± SD, n = 4 technical replicates) with corresponding mean Bliss score plots (bottom, n = 4 technical replicates) showing the scores from each combination's concentration matrix and colored as a heat map (blue indicates synergy; yellow indicates additivity; red indicates antagonism). Combinations of the DNA-damaging drug topotecan with the ATR inhibitors berzosertib (A) and elimusertib (B). Combinations of the DNA-damaging drug trabectedin with either berzosertib (C) or elimusertib (D). Data are shown for complex tumor spheroids grown with the malignant cell lines (from left): 287954-098-R-J1 (Ewing sarcoma), NCI-H841 (SCLC), DMS 114 (SCLC), COR L88 (SCLC), and 425362-245-T-J1 (melanoma).

Figure S2. Heat maps of Bliss synergy scores across the combination dose-response matrices for all twenty-six cell lines grown as multicellular complex spheroids exposed to each DNA damaging agent (A, TMZ; B, topotecan; C, trabectedin) in combination with PARP inhibitors, olaparib or talazoparib.

Combination of PARP inhibitors and temozolomide. A, Mean Bliss matrix scores were calculated from each combination's concentration matrix [(5 concentrations of drug A) × (6 concentrations of drug B) = (30 combination concentrations)]. The scores (n = 546) are graphed for all combinations of DNA-damaging drugs with DNA repair inhibitors tested (n = 21) in all complex tumor spheroid models (n = 26). B, The same data shown in A) but combinations of the DNA-damaging drug temozolomide with either PARP inhibitor olaparib or talazoparib are highlighted in blue (n = 52), whereas all other combinations are shown in light gray. C, A scatter plot of the mean Bliss matrix scores from the temozolomide combinations with olaparib and talazoparib (Pearson r = 0.73, two-tailed P < 0.0001). Concentration–response curves (top, mean ± SD, n = 4 technical replicates) from combinations of temozolomide with either olaparib (D) or talazoparib (E) and corresponding mean Bliss score plots (bottom, n = 4 technical replicates) showing the scores from each combination's concentration matrix and colored as a heat map (blue indicates synergy; yellow indicates additivity; red indicates antagonism). Data are shown for complex tumor spheroids grown with the malignant cell lines (from left): VA-ES-VJ (epithelioid sarcoma), SYO-1 (synovial sarcoma), 287954-098-R-J1 (Ewing sarcoma), 349418-098-R (NSCLC), and DMS 114 (SCLC).

Figure S3. Heat maps of Bliss synergy scores across the combination dose-response matrices for all twenty-six cell lines grown as multicellular complex spheroids exposed to each DNA damaging agent (A, TMZ; B, topotecan; C, trabectedin) in combination with the ATM inhibitor, AZD1390.

Objectives: The efficacy of treatments for advanced, recurrent metastatic endometrial cancers (EC) remains limited. In-depth, high-dimensional immune profiling of recurrent EC subtypes is much needed in order to describe the immune composition within the tumor microenvironment. Methods: A 2:1 randomized phase 2 trial was conducted comparing the combination of cabozantinib and nivolumab (Arm A) versus nivolumab (Arm B) in women with recurrent measurable EC (NCT03367741). A third exploratory cohort (Arm C) included EC patients with the carcinosarcoma subtype, and patients who had prior immunotherapy (IO). Fresh baseline core biopsies were collected and processed into single-cell suspensions for high-dimensional CyTOF analysis with a 36-marker immune profiling panel. Immune composition was determined by unsupervised single-cell clustering using PhenoGraph and visualized by uniform manifold approximation and projection (UMAP). The differential abundance (DA) of unique immune cell subsets present in the baseline biopsies was determined using the diffcyt-DA-EdgeR method. Results: Cabozantinib plus nivolumab demonstrates improved PFS and ORR compared to nivolumab in heavily pre-treated women with recurrent EC. Unsupervised clustering of CD45+ cells from baseline biopsies (n=40) using PhenoGraph resulted in 35 unique immune cell subsets defined their expression of lineage and activation/checkpoint markers (e.g. PD-1, PD-L1, TIGIT, 4-1BB, CD39, CD69 and others). Broadly, these PhenoGraph-defined subsets constitute the major immune populations including CD4 and CD8 T cells, regulatory T cells, γδ T cells, B cells, innate lymphoid/natural killer (NK) cells, and monocyte/macrophages and dendritic cells. In-depth CyTOF analysis of the tumor microenvironment identified a higher proportion of activated tissue-resident γδ T cells in patients who had prior IO and who benefit from the combination therapy (non-progressors, n=4; progressors, n=5; 6 log fold-change, adjusted P=0.001). We did not observe any statistically significant differences in the baseline immune composition between IO-naïve (Arm A and B) endometrioid (n=13), serous (n=7) and carcinosarcoma tumors (n=4). However, in comparison to endometrioid and serous tumors, carcinosarcoma tumors may trend towards a lower abundance (not significant) of CD45RA+CD27+CD28+ CD4 and CD8 T cells, and CD45RA+CD69- NK cells. All three histological subtypes had a similar abundance of activated, tissue-resident PD-1+TIGIT+CD69+CD103+ CD8 T cells. Microsatellite instability high (MSI-H; n=2; both endometrioid) tumors were excluded from our analysis. Conclusions: In comparison to nivolumab monotherapy, combination cabozantinib and nivolumab treatment improved PFS and ORR in recurrent EC patients. To our knowledge, this is the first study reporting in-depth immune profiling analysis of core biopsies from EC patients. The tumor immune microenvironment of EC is comprised of several immune populations with unique phenotypes related to their expression of activation and checkpoint markers. The efficacy of treatments for advanced, recurrent metastatic endometrial cancers (EC) remains limited. In-depth, high-dimensional immune profiling of recurrent EC subtypes is much needed in order to describe the immune composition within the tumor microenvironment. A 2:1 randomized phase 2 trial was conducted comparing the combination of cabozantinib and nivolumab (Arm A) versus nivolumab (Arm B) in women with recurrent measurable EC (NCT03367741). A third exploratory cohort (Arm C) included EC patients with the carcinosarcoma subtype, and patients who had prior immunotherapy (IO). Fresh baseline core biopsies were collected and processed into single-cell suspensions for high-dimensional CyTOF analysis with a 36-marker immune profiling panel. Immune composition was determined by unsupervised single-cell clustering using PhenoGraph and visualized by uniform manifold approximation and projection (UMAP). The differential abundance (DA) of unique immune cell subsets present in the baseline biopsies was determined using the diffcyt-DA-EdgeR method. Cabozantinib plus nivolumab demonstrates improved PFS and ORR compared to nivolumab in heavily pre-treated women with recurrent EC. Unsupervised clustering of CD45+ cells from baseline biopsies (n=40) using PhenoGraph resulted in 35 unique immune cell subsets defined their expression of lineage and activation/checkpoint markers (e.g. PD-1, PD-L1, TIGIT, 4-1BB, CD39, CD69 and others). Broadly, these PhenoGraph-defined subsets constitute the major immune populations including CD4 and CD8 T cells, regulatory T cells, γδ T cells, B cells, innate lymphoid/natural killer (NK) cells, and monocyte/macrophages and dendritic cells. In-depth CyTOF analysis of the tumor microenvironment identified a higher proportion of activated tissue-resident γδ T cells in patients who had prior IO and who benefit from the combination therapy (non-progressors, n=4; progressors, n=5; 6 log fold-change, adjusted P=0.001). We did not observe any statistically significant differences in the baseline immune composition between IO-naïve (Arm A and B) endometrioid (n=13), serous (n=7) and carcinosarcoma tumors (n=4). However, in comparison to endometrioid and serous tumors, carcinosarcoma tumors may trend towards a lower abundance (not significant) of CD45RA+CD27+CD28+ CD4 and CD8 T cells, and CD45RA+CD69- NK cells. All three histological subtypes had a similar abundance of activated, tissue-resident PD-1+TIGIT+CD69+CD103+ CD8 T cells. Microsatellite instability high (MSI-H; n=2; both endometrioid) tumors were excluded from our analysis. In comparison to nivolumab monotherapy, combination cabozantinib and nivolumab treatment improved PFS and ORR in recurrent EC patients. To our knowledge, this is the first study reporting in-depth immune profiling analysis of core biopsies from EC patients. The tumor immune microenvironment of EC is comprised of several immune populations with unique phenotypes related to their expression of activation and checkpoint markers.

Background The inhibition of insulin-like growth factor receptor-1 (IGF-1R) induces cell cycle arrest and enhancing the effect of castration by delay of progression of human prostate cancer models. Linsitinib is a small molecule and potent dual inhibitor of IGF-1R and insulin receptor tyrosine kinase activity. We report results of a single-arm, phase II study evaluating the safety and efficacy of linsitinib in men with chemotherapy-naïve asymptomatic or mildly symptomatic metastatic castration resistant prostate cancer (mCRPC). Methods Patients received at 150 mg orally twice daily on a 28-day cycle. The primary endpoint was prostate specific (PSA) response at 12 weeks and correlative studies included circulating tumor cells (CTCs) and circulating endothelial cells (CECs). Results Seventeen patients, median age 68 (55–78) and pre-treatment PSA of 55.23 (2.46–277.60) were enrolled and completed 12 weeks of therapy. All but two patients discontinued therapy secondary to PSA progression, which met the predefined futility criteria and led to early termination of this study. Overall best response (RECIST v1.1) included a partial response in 1 patient and stable disease in 8 patients. Higher baseline CTCs were associated with higher pre-treatment PSA levels (Spearman r = 0.49, p = 0.04) but no correlation between PSA progression and CTCs/CECs were observed. Most common adverse events included fatigue, nausea/vomiting, AST/ALT changes and prolonged QT interval. Conclusions Single-agent linsitinib was safe and well tolerated but failed to show activity in men with mCRPC. These results highlight the complexity of using IGF-1R as a therapeutic target in this patient population. ClinicalTrials.gov NCT01533246.