Search

Antibody-drug conjugates (ADC) leverage the specificity of antibodies to direct the delivery of potent cytotoxic agents to cancer cells. To fully leverage this specificity, targets with significant differential tumor versus normal tissue expression are ideal for ADC development. Placental alkaline phosphatases, ALPP and ALPPL2, are proteins present during fetal development but are found in several tumor types, making them an attractive ADC target. These cell membrane-attached phosphatases form homo- and heterodimers and play key roles in nucleotide recycling. SGN-ALPV is a novel investigational vedotin ADC comprised of a humanized IgG1 monoclonal antibody conjugated to the microtubule disrupting agent monomethyl auristatin E (MMAE) via a protease-cleavable peptide linker that has been clinically validated in multiple ADC programs. Here we characterize the target antigens, ALPP and ALPPL2, and evaluate SGN-ALPV antitumor activity in preclinical models. Immunohistochemistry and genomic data mining showed that ALPP and ALPPL2 have a highly restricted normal tissue expression yet high expression in several solid tumor types including ovarian, endometrial, germ cell, non-small cell lung and gastric carcinomas. Importantly, normal tissue expression is restricted to placenta and reproductive tissue, with low levels in lung tissue. SGN-ALPV utilizes a humanized antibody, h12F3, that is highly specific for both human and cynomolgus monkey ALPP and ALPPL2 proteins, but not other related phosphatases. In vitro, upon binding to SGN APLV, the antigens, ALPP and ALPPL2, are internalized to lysosomal vesicles releasing the MMAE payload, whose cytotoxic features drive mitotic arrest, apoptosis, and the induction of immunogenic cell death. Additionally, SGN-ALPV mediates antibody dependent cellular cytotoxicity (ADCC) and antibody-dependent cellular phagocytosis (ADCP) in vitro but lacks complement-dependent cytotoxicity. In preclinical studies, SGN-ALPV exhibits robust antitumor activity in cell line- and patient-derived xenograft models of ovarian, lung, pancreatic, and gastric carcinoma including models with both homogenous and heterogeneous expression of ALPP and ALPPL2, consistent with robust bystander activity of vedotin ADCs. Although, the contribution of ADCC and ADCP in vivo is currently unknown, it is expected that the antitumor activity of SGN-ALPV is mediated by MMAE cytotoxicity. SGN-ALPV was well tolerated in non-human primates (NHP) and exhibited linear pharmacokinetic characteristics, with a toxicity profile consistent with other vedotin-based ADCs. In summary, differential expression of ALPP and ALPPL2 in the tumor versus normal tissue, antibody specificity, antitumor activity, and tolerability of SGN-ALPV provide a strong rationale for the initiation of a planned first-in-human Phase 1 clinical study. Citation Format: Sarah Anderson, Nanna Hansen, Robert Lawrence, Aroon T. Chande, David Ortiz, Christopher Carosino, Esther Trueblood, Nicole Stevens, Kerry Klussman, Angela Epp, Bill Arthur, Shyra Gardai, Hector Rincon. SGN-ALPV a novel, investigational vedotin ADC demonstrates highly effective targeting of oncofetal phosphatases ALPP and ALPPL2 in preclinical models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1766.

NL-201 is a computationally-designed IL-2/IL-15 cytokine mimetic under investigation in a phase 1 clinical trial for patients with solid tumors. This de novo protein signals through the IL-2Rβ/ɣ complex in an IL-2Rα-independent fashion, supporting immune function of both T cells and natural killer (NK) cells. In syngeneic tumor models of solid tumors, NL-201 demonstrates marked antitumor activity through multiple mechanisms of action, including upregulation of granzyme B in cytotoxic CD8+ T cells and NK cells. Here, we further explored the therapeutic potential of NL-201 in preclinical models of B cell malignancies. In cell-based studies, NL-201 augmented antibody-dependent cellular cytotoxicity (ADCC) of Toledo DLBCL (diffuse large B cell lymphoma) cells incubated with human NK cells and CD20 antibodies. Importantly, direct activation by NL-201 was specific to NK cells and did not impact STAT5 phosphorylation nor proliferation of the tested malignant cells. The administration of NL-201 monotherapy in mice bearing A20 B cell lymphoma syngeneic tumors yielded robust inhibition of tumor growth. Tumor xenografts from implanted Pfeiffer cells have also been evaluated to understand the activity of NL-201 in a model closely resembling DLBCL physiology. In summary, NL-201 is active as a single agent and in combination with CD20-targeted therapies in vitro, as well as demonstrating monotherapy activity in in vivo preclinical models of B cell malignancies as well as in combination with checkpoint inhibitors. The NL-201 enhancement of cytotoxicity by CD8+ T cells and NK cells supports additional exploration aimed at optimizing combinatorial approaches. These findings will inform the design of planned clinical studies with NL-201 in patients with B-cell lymphomas and other hematological malignancies. Citation Format: Justin Huard, Christie Mortales, Alex Chen, Ryan Swanson, Carl Walkey, Bill Arthur. NL-201, a de novo agonist of IL-2 and IL-15 receptors, demonstrates antitumor activity in preclinical B cell lymphoma models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 4227.

The RAS-MAPK pathway controls many cellular programs, including cell proliferation, differentiation, and apoptosis. In colorectal cancers, recurrent mutations in this pathway often lead to increased cell signaling that may contribute to the development of neoplasms, thereby making this pathway attractive for therapeutic intervention. To this end, we developed a 26-member gene signature of RAS-MAPK pathway activity utilizing the Affymetrix QuantiGene Plex 2.0 reagent system and performed both primary and confirmatory gene expression-based high-throughput screens (GE-HTSs) using KRAS mutant colon cancer cells (SW837) and leveraging a highly annotated chemical library. The screen achieved a hit rate of 1.4% and was able to enrich for hit compounds that target RAS-MAPK pathway members such as MEK and EGFR. Sensitivity and selectivity performance measurements were 0.84 and 1.00, respectively, indicating high true-positive and true-negative rates. Active compounds from the primary screen were confirmed in a dose-response GE-HTS assay, a GE-HTS assay using 14 additional cancer cell lines, and an in vitro colony formation assay. Altogether, our data suggest that this GE-HTS assay will be useful for larger unbiased chemical screens to identify novel compounds and mechanisms that may modulate the RAS-MAPK pathway.

Combination therapies are a cornerstone of cancer treatment, enabling a strategy of engaging multiple mechanisms of action. The in vitro testing of drug combinations is therefore a key step in drug discovery, and plate-based assays and robotics now allow for high-throughput cytotoxicity experiments that test combinations at doses diluted at high resolution. While additivity models such as Bliss, Loewe and Highest Single Agent (HSA) are well-established for gauging the synergy or antagonism of combinations, new frameworks are required for incorporating these models in the context of hundreds of experiments testing dozens of dose combinations per experiment. We have developed a framework that compares observed cytotoxicity results against the predictions from additivity models on a dose-by-dose basis across an entire matrix of dose combinations. Visualizations readily highlight any synergistic or antagonistic deviations from additivity. Statistical tests are then performed using 3x3 blocks of adjacent combinations. Such an approach mutes the effects of outliers and rewards consistency of activity in nearby doses. To capture the overall trend in an experiment and to facilitate comparisons across multiple experiments, we propose a summary metric, TSDA (total significant deviation from additivity) that takes into account both the absolute deviations from additivity and their statistical significance. Generally, positive values of TSDA indicate overall synergy, while negative values indicate overall antagonism. Null distributions for TSDA have been generated using simulations of sham experiments (same drug for both axes in a combination). We have applied this framework to over 1500 individual combination experiments spanning a range of cell lines and therapeutic mechanisms. The framework has been used to successfully guide follow-up in vivo experiments and to inform clinical plans. Citation Format: Robert Thurman, Bill Arthur. Novel framework for quantifying synergy in high-throughput drug combination cytotoxicity experiments [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 835.

The prognosis of acute myeloid leukemia (AML) is very poor, with overall survival rates of approximately 50% in children and 25% in adults, despite reaching the limits of conventional cytotoxic chemotherapy. Therapeutic targeting in AML through either surface antigens or aberrantly activated cell signaling pathways are two emerging targeted treatment strategies in efforts to further improve outcomes. FLT3/ITD mutations are among the most common somatic mutations in AML and confer an inferior prognosis. Outcomes for this group of patients have recently been shown to be improved with targeting mutant FLT3 with FLT3 inhibitors (FLT3i). Patients with FLT3/ITD mutant AML express high levels of CD33, and the benefit of CD33 targeted agents in this group of patients has been demonstrated. Thus, we sought to investigate the combination strategy of FLT3i with SGN-CD33A (vadastuximab tarline), a CD33-directed antibody drug conjugate (ADC), in FLT3/ITD AML. Combination experiments with 6 FLT3 inhibitors (midostaurin, quizartinib, crenolanib, gilteritinib, AMG925, G-749) and SGN-CD33A were performed in 2 FLT3/ITD positive cell lines (MV4-11, MOLM-13). Five non-FLT3 mutant cell lines (Kasumi1, KG-1, ME-1, TF1, THP1) as well as normal hematopoietic cells (CD34+ HSCs, megakaryocytes) with and without serum were used as controls. In each pairing, a total of 240 individual drug combinations of SGN-CD33A and FLT3i were evaluated at doses spanning 4 and 6 log-folds respectively. Across the matrix of dose combinations, we compared observed cytotoxicity against 3 standard additivity models (Loewe, Bliss, and Highest Single Agent (HSA)) and summarized each experiment using metrics gauging the degree of synergy or antagonism observed based on those additive models. The combination was also evaluated with xenograft experiments. Following transplantation with MOLM-13 cells, mice received no treatment or treatment with FLT3i, SGN-CD33A, or the combination and survival was evaluated using Kaplan-Meier methods. We performed a total of 88 in vitro cell line experiments, 25 in FLT3/ITD positive cell lines. Aggregate statistics over all dose combinations showed significant synergistic effects of the SGN-CD33A/FLT3i combinations in FLT3/ITD mutant cells. Importantly, no synergistic effects were observed in any of the non-mutant cell lines. Moreover, in most of the experiments in FLT3/ITD-positive cell lines the dose combination producing the highest synergistic effect was within one log of the IC50 for both SGN-CD33A and the FLT3i (Fig 1A). These findings indicate that synergy was achieved at clinically meaningful doses. We subsequently evaluated the effect of the combination in vivo and following transplantation of MOLM-13 cells, xenografts were treated with either quizartinib alone, SGN-CD33A alone, quizartinib+SGN-CD33A, or no treatment. Subjects treated with SGN-CD33A/FLT3i combination had improved survival compared to the untreated or single agent mice (Figure 1B). Patients with FLT3/ITD mutations have a poor prognosis, despite intensive therapy including hematopoietic stem cell transplant. Importantly, this group of patients harbors 2 potential therapeutic targets, namely aberrant FLT3 activation and high CD33 surface expression. We have shown combining FLT3 inhibitors with the CD33 directed ADC SGN-CD33A results in a synergistic effect with increased cell killing and prolonged survival in xenografts. This synergistic effect was most pronounced at doses on the lower range of SGN-CD33A, suggesting that a lower dose of CD33-directed ADCs could be used thus mitigating excess hematopoietic toxicity. These treatments have little overlap in toxicity profiles suggesting that they can be safely combined in patients, and further clinical investigation of this strategy is warranted. Disclosures Thurman: Seattle Genetics: Employment, Equity Ownership. Rohm:Seattle Genetics: Employment, Equity Ownership. Biechele:Seattle Genetics: Employment, Equity Ownership. Arthur:Seattle Genetics: Employment, Equity Ownership.

The cell surface antigen CD33 is expressed on the majority of AML blasts and is appropriate for immunotherapeutic targeting with antibody drug conjugates (ADCs). Expression of CD33 is in part mediated by splicing of the CD33 transcript, and has been demonstrated to be one of the factors that may mediate response to the ADC gemtuzumab ozogamicin, which results in significant benefit in some patients but lacks responses in others. Splicing of the CD33 transcript is in part regulated by a single nucleotide polymorphism (SNP) in exon 2 (e2) that causes a C>T substitution and the resultant skipping of e2. CD33 thus exists in 2 main isoforms, as either a full length (FL) transcript or a truncated version missing e2 (Δe2), which includes the IgV binding domain that is the epitope for diagnostic and therapeutic antibodies (Ab). The CC genotype encodes the FL isoform at a higher rate compared to the CT or TT, and the TT genotype encodes the short isoform at a higher rate compared to CT or CC. SGN-CD33A is a CD33-directed ADC, utilizing a pyrolobenzodiazepine (PBD) dimer. SGN-CD33A has been evaluated in multiple clinical trials as either monotherapy or in combination. We hypothesized that the patient's CD33 genotype would impact CD33 expression as well as response characteristics following treatment with SGN-CD33A. We analyzed CD33 genotype variation in bone marrow (BM) or peripheral blood (PB) samples from patients treated with SGN-33A as either monotherapy (NCT01902329; n=133) or in combination with hypomethylating agents (HMAs; NCT02785900; n=83). CD33 SNP genotyping was determined on gDNA using RFLP PCR with 2 restriction enzymes recognizing cut sites generated by the C and T polymorphisms and genotype confirmed using fragment length analysis (CC=108, CT=86, TT=22). CD33 surface expression on the AML blasts was determined by flow cytometric analysis using the human anti-CD33 monoclonal Abs HIM3-4 and H212, which bind to the membrane-proximal C2-set and V-set domain, respectively. HIM-34 measured CD33 levels independent of SNP-driven splice variation. The h2H12 epitope is within e2, thus its binding may be susceptible to splice variation. We subsequently evaluated the association of CD33 genotype with pharmacokinetic (PK), clinical and other variables using a generalized regression model. Patients classified as CC genotype had significantly higher surface CD33 expression as determined by flow cytometry in both BM and PB. In accordance with observed differences in CD33 expression, we also found drug exposure demonstrated an inverse relationship according to CC genotype in both mono and combination therapy trials. For monotherapy, compared to patients with CC and CT genotypes, patients with TT genotypes had significantly higher drug exposure following SGN-CD33A. Patients with TT had higher AUCs following the first and last doses of SGN-33A (p < 10-4 -; Fig 1). Cmax following SGN-CD33A exposure was higher in TT genotype patients compared to the CT and CC (p< 10-1.5 for Cmax following the first dose and p We next examined expression and subcellular localization of CD33 to elucidate the mechanism by CD33 variation contributes to cell surface presentation. Transfection of cDNA encoding the FL CD33 transcript resulted in increased cell surface expression, as indicated by flow cytometry with both HIM3-4 and h2H12. In contrast, both Abs failed to detect cell surface CD33 following transfection with cDNA encoding the Δe2 variant. Examination of the intracellular compartment revealed that HIM3-4, but not 2H12, binds to the Δe2 variant in a pattern localized proximal to the nucleus. Taken together, our findings suggest that the Δe2 splice CD33 variant lacks the portion of the V-set domain required for h2H12/SGN-CD33A binding and does not efficiently traffic to the cell surface. We show that CD33 SNP genotype is associated with CD33 expression, with CC patients demonstrating higher CD33 as detected by flow cytometry; and that CD33 SNP genotype affects the PK profile of SGN-CD33A, with TT patients having higher levels of drug exposure. Our findings suggest that the CD33 genotype can impact CD33 expression, PK profile, and trafficking of bound agents and thus may impact therapeutic targeting of CD33-directed agents. Disclosures Wang: Seattle Genetics: Employment, Equity Ownership. Rohm:Seattle Genetics: Employment, Equity Ownership. Biechele:Seattle Genetics: Employment, Equity Ownership. Means:Seattle Genetics: Employment, Equity Ownership. Thurman:Seattle Genetics: Employment, Equity Ownership. Arthur:Seattle Genetics: Employment, Equity Ownership.

The AMDTreat computer program ( . Treatment with caustic chemicals typically is used to increase pH and decrease concentrations of dissolved aluminum, iron, and/or manganese in largevolume, metal-laden discharges from active coal mines. Generally, aluminum and iron can be removed effectively at near-neutral pH (6 to 8), whereas active manganese removal requires treatment to alkaline pH (~10). The treatment cost depends on the specific chemical used (NaOH, CaO, Ca(OH)2, Na2CO3, or NH3) and increases with the quantities of chemical added and sludge produced. The pH and metals concentrations do not change linearly with the amount of chemical added. Consequently, the amount of caustic chemical needed to achieve a target pH and the corresponding effluent composition and sludge volume can not be accurately determined without empirical titration data or the application of geochemical models to simulate the titration of the discharge water with caustic chemical(s). http://amd.osmre.gov/ ) is widely used to compute costs for treatment of coal-mine drainage. Although AMDTreat can use results of empirical titration with industrial grade caustic chemicals to compute chemical costs for treatment of net-acidic or net-alkaline mine drainage, such data are rarely available. To improve the capability of AMDTreat to estimate (1) the quantity and cost of caustic chemicals to attain a target pH, (2) the concentrations of dissolved metals in treated effluent, and (3) the volume of sludge produced by the treatment, a titration simulation is being developed using the geochemical program PHREEQC (wwwbrr.cr.usgs.gov/projects/GWC_coupled/phreeqc/) that will be coupled as a module to AMDTreat. The simulated titration results can be compared with or used in place of empirical titration data to estimate chemical quantities and costs. This paper describes the development, evaluation, and potential utilization of the PHREEQC titration module for AMDTreat.

Antibody-drug conjugates (ADC) deliver cytotoxic payloads to target cells via receptor-mediated internalization. We have demonstrated that SGN-35 (cAC10-vcMMAE) binds to CD30 on target cells and releases intracellular monomethyl auristatin E (MMAE) in vitro, contributing to potent killing of CD30+ cells. Using a set of auristatin-based ADCs, we evaluated whether the intratumoral concentration of the released payload is a predictive indicator of antitumor activity in preclinical models. We first treated CD30+, anaplastic large cell lymphoma (ALCL) Karpas 299 tumors with SGN-35 or a non-binding control ADC (IgG-vcMMAE). The plasma ADC and MMAE PK measurements were similar in SGN-35 and IgG-vcMMAE treated tumor-bearing animals. In contrast, SGN-35 generated 5-fold greater intratumoral MMAE exposure than IgG-vcMMAE, which correlated with tumor regression observed in the SGN-35- treated group. We then examined the relationship between released MMAE and in vitro cytotoxicity by targeting different receptors expressed on the same cells. The lymphoma cell line L82 expresses CD30, CD70, and CD71 receptors. Although auristatin-based ADCs targeting each of these targets has a different IC50, the intracellular MMAE was similar when 50% growth inhibition was achieved (approximately 200 nM). In vivo, CD30, CD70, or CD71-directed ADCs demonstrated similar antitumor activity for a given dose level (0.5, 1, or 3 mg/kg). Concordantly, equal intratumoral MMAE concentrations were observed for a given dose level, regardless of the antigen target. The cell permeable MMAE has been shown to mediate bystander killing in vitro. To evaluate whether intratumoral MMAE mediates bystander killing in vivo, an admixed tumor model consisting CD30+ and CD30- Karpas 299 cells was treated with SGN-35. As predicted, both CD30+ and CD30- tumor cells were killed, presumably due to intratumoral diffusion of MMAE across cell membranes. In contrast, a conjugate with a highly polar form of auristatin (MMAF) was unable to kill the CD30-negative cell population of the admixed tumors. These preclinical results provide evidence that the concentration of released MMAE within the tumor correlates with the ADC antitumor activity in vitro and in vivo. Moreover, the in vivo bystander activity of MMAE may improve utility of this chemotype in tumors with heterogeneous antigen expression. Citation Format: Fu Li, Xinqun Zhang, Kim Emmerton, Mechthild Jonas, Jocelyn Setter, Bill Arthur, Nicole Okeley, Robert Lyon, Dennis Benjamin, Che-Leung Law. Relationship between in vivo antitumor activity of ADC and payload release in preclinical models. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3694. doi:10.1158/1538-7445.AM2014-3694

We have tested PI3K and MAPK targeting compounds in a panel of 69 breast cancer cell lines. Following the molecular stratification of breast cancer tumors, we have used ER and HER2 biomarkers to stratify cell lines into HER2 positive (HER2+), ER positive (ER+), and triple negative (TN; ER-, PR- and HER2-) subgroups. Both gene expression and response to compounds showed significant differences among the subtypes. Response to mTOR and AKT inhibition was the greatest in ER+ and HER2+ groups (86%, 85% for mTORi and 64%, 77% for AKTi response rates, respectively), whereas TN cell lines showed preferential sensitivity to MEK inhibition (30%). The combination of AKT plus MEK and mTOR plus MEK inhibitors showed synergy in TN tumors (56% and 80% response rates, respectively), particularly in the epithelial subgroup. We believe that RAS pathway deregulation, as reported by a gene expression signature of RAS activation, is the key latent variable that explains the difference in responses. This conclusion is based on pathway activation differences between subtypes. We observe that the TN group has both elevated PI3K and MAPK signaling, evident by high levels of RAS, pERK, and pMEK signatures, whereas the ER+/GGI+ group only has high levels of PI3K signaling. Thus, mTOR, AKT, or MEK alone is insufficient to inhibit the TN lines, whereas the combos are very effective. Finally, we have assessed the prevalence of the EMT and RAS signatures across subtypes of breast tumors. Similar to cell lines, the RAS signature is markedly higher in TN compared to ER+/GGI+ and, to a lesser extent, compared to HER2+ tumors. The EMT signature on the other hand is not significantly different among subtypes. This would suggest that the response patterns observed in ER+/GGI+ and TN/E subtypes of this breast panel will be applicable to a substantial subpopulation of breast tumors (∼38%). Importantly, these breast subtypes represent the largest unmet need and are a high priority for the clinical oncology. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 1986.