Anti-drug Responses Research Articles

Graph-pMHC: graph neural network approach to MHC class II peptide presentation and antibody immunogenicity.

Antigen presentation on MHC class II (pMHCII presentation) plays an essential role in the adaptive immune response to extracellular pathogens and cancerous cells. But it can also reduce the efficacy of large-molecule drugs by triggering an anti-drug response. Significant progress has been made in pMHCII presentation modeling due to the collection of large-scale pMHC mass spectrometry datasets (ligandomes) and advances in machine learning. Here, we develop graph-pMHC, a graph neural network approach to predict pMHCII presentation. We derive adjacency matrices for pMHCII using Alphafold2-multimer and address the peptide-MHC binding groove alignment problem with a simple graph enumeration strategy. We demonstrate that graph-pMHC dramatically outperforms methods with suboptimal inductive biases, such as the multilayer-perceptron-based NetMHCIIpan-4.0 (+20.17% absolute average precision). Finally, we create an antibody drug immunogenicity dataset from clinical trial data and develop a method for measuring anti-antibody immunogenicity risk using pMHCII presentation models. Our model increases receiver operating characteristic curve (ROC)-area under the ROC curve (AUC) by 2.57% compared to just filtering peptides by hits in OASis alone for predicting antibody drug immunogenicity.

Do Personality Traits Matter? Exploring Anti-drug Behavioral Patterns in a Computer-Assisted Situated Learning Environment

Drug abuse has been and continues to be, a common social issue worldwide, yet the efficiency of widely adopted sweeping speech for anti-drug campaigns has proven inefficient. To provide students with a safe and efficient learning situation related to drug refusal skills, we used a novel approach rooted in a serious learning game and concept map during a brief extracurricular period to help students understand drugs and their negative effects. The proposed game-based situational learning system allowed all students to participate simultaneously and individually in multiple scenarios of drug temptation posed by peers and classmates to practice responding and refusing drugs in school and community settings. Moreover, to explore whether different personality traits (such as the Big Five personality traits) result in different anti-drug responses, we used a serious game to conduct an anti-drug experiment on 53 junior middle school students aged 13–15. Each participant’s decision-making process was recorded in the serious game as behavioral patterns for lag sequential analysis (LSA). The outcomes revealed seven behavioral patterns including differentiation (D), acceptance (A), effective (ER) and ineffective responses (IR), effective (ES) and ineffective solution-seeking (IS), and failure to refuse (F). The GSEQ (Generalized Sequential Querier) which is a computer program for analyzing sequential observational data was used. The results indicated the following: (1) Neuroticism was performed at a relatively low level under the guidance of a concept map. (2) “Neuroticism” was associated with the lowest risk of accepting drugs. (3) Students with “openness to experiences” were at high risk of accepting drugs. (4) Almost all personality behavioral transition diagrams showed that failure to refuse (F) drugs was followed by inefficient seeking of help (SI) and inefficient refusal (RI). These findings provide reference points for designing adaptive anti-drug education programs.

Drug-specific Treg cells are induced during desensitization procedure for rituximab and tocilizumab in patients with anaphylaxis

Biologic agents (BA) are able to induce an adaptive immune response in a proportion of exposed patients with the onset of anti-drug antibodies (ADA), which are usually responsible for hypersensitivity reactions (HR). Drug desensitization (DD) for BA allows transient clinical tolerance to the drug in reactive patients. The paper aimed to analyse the modification of drug-specific immune responses along DD in two patients with previous ADA-mediated HR (anaphylaxis) to rituximab and tocilizumab. The in vivo and in vitro assays of humoral and cellular response to drugs were carried out in a longitudinal manner throughout the DD cycles. We observed a progressive decrease of the pre-procedure ADA titer with negativization during the DD cycles in both patients. The monitoring of the drug-specific effector cell response showed the decrease in the BA-induced proliferation, while T cell response to unrelated antigens resulted unmodified along the DD cycles. Lastly, the increase of circulating drug-specific Treg cells mainly producing IL-35 were shown during the DD treatment. This study provides evidence that DD treatment to two BA inhibits humoral and cellular anti-drug response by increasing regulatory T cells and cytokines in an antigen-restricted manner. These modifications could contribute to the safety of the procedure.

Predicting Anticancer Drug Response With Deep Learning Constrained by Signaling Pathways.

Thanks to the availability of multiomics data of individual cancer patients, precision medicine or personalized medicine is becoming a promising treatment for individual cancer patients. However, the association patterns, that is, the mechanism of response (MoR) between large-scale multiomics features and drug response are complex and heterogeneous and remain unclear. Although there are existing computational models for predicting drug response using the high-dimensional multiomics features, it remains challenging to uncover the complex molecular mechanism of drug responses. To reduce the number of predictors/features and make the model more interpretable, in this study, 46 signaling pathways were used to build a deep learning model constrained by signaling pathways, consDeepSignaling, for anti–drug response prediction. Multiomics data, like gene expression and copy number variation, of individual genes can be integrated naturally in this model. The signaling pathway–constrained deep learning model was evaluated using the multiomics data of ∼1000 cancer cell lines in the Broad Institute Cancer Cell Line Encyclopedia (CCLE) database and the corresponding drug–cancer cell line response data set in the Genomics of Drug Sensitivity in Cancer (GDSC) database. The evaluation results showed that the proposed model outperformed the existing deep neural network models. Also, the model interpretation analysis indicated the distinctive patterns of importance of signaling pathways in anticancer drug response prediction.

Detecting and predicting neutralization of alemtuzumab responses in MS.

ObjectiveTo test the hypothesis that antidrug antibodies (ADAs) against alemtuzumab could become relevant after repeated treatments for some individuals, possibly explaining occasional treatment resistance.MethodsRecombinant alemtuzumab single-chain variable fragment antibody with a dual tandem nanoluciferase reporter linker was made and used to detect binding ADAs. Alemtuzumab immunoglobulin G Alexa Fluor 488 conjugate was used in a competitive binding cell-based assay to detect neutralizing ADAs. The assays were used to retrospectively screen, blinded, banked serum samples from people with MS (n = 32) who had received 3 or more cycles of alemtuzumab. Lymphocyte depletion was measured between baseline and about 1 month postinfusion.ResultsThe number of individuals showing limited depletion of lymphocytes increased with the number of treatment cycles. Lack of depletion was also a poor prognostic feature for future disease activity. ADA responses were detected in 29/32 (90.6%) individuals. Neutralizing antibodies occurred before the development of limited depletion in 6/7 individuals (18.8% of the whole sample). Preinfusion, ADA levels predicted limited, postinfusion lymphocyte depletion.ConclusionsAlthough ADAs to alemtuzumab have been portrayed as being of no clinical significance, alemtuzumab-specific antibodies appear to be clinically relevant for some individuals, although causation remains to be established. Monitoring of lymphocyte depletion and the antidrug response may be of practical value in patients requiring additional cycles of alemtuzumab. ADA detection may help to inform on retreatment or switching to another treatment.

The Irony of Humanization: Alemtuzumab, the First, But One of the Most Immunogenic, Humanized Monoclonal Antibodies.

Alemtuzumab was designed to reduce the immunogenicity of the parent CD52-specific rat immunoglobulin. Although originally marketed for use in cancer (Mabcampath®), alemtuzumab is currently licensed and formulated for the treatment of relapsing multiple sclerosis (Lemtrada®). Perhaps due to its history as the first humanized antibody, the potential of immunogenicity of the molecule has been considered inconsequential, and anti-drug antibodies (ADA) responses were similarly reported as being clinically insignificant. Nonetheless, despite humanization and depletion of peripheral T and B cells, alemtuzumab probably generates the highest frequency of binding and neutralizing ADA of all humanized antibodies currently in clinical use, and they occur rapidly in a large majority of people with MS (pwMS) on alemtuzumab treatment. These ADA appear to be an inherent issue of the biology of the molecule—and more importantly, the target—such that avoidance of immunogenicity-related effects has been facilitated by the dosing schedule used in clinical practice. At the population level this enables the drug to work in most pwMS, but in some individuals, as we show here, antibody neutralization appears to be sufficiently severe to reduce efficacy and allow disease breakthrough. It is therefore imperative that efficacy of lymphocyte depletion and the anti-drug response is monitored in people requiring additional cycles of treatment, notably following disease breakthrough. This may help inform whether to re-treat or to switch to another disease-modifying treatment.

Induction and Impact of Anti-Drug Responses Elicited by a Human Recombinant Coagulation Factor FXaI16L in Preclinical Species.

This paper presents a systemic investigation of ADA development and ADA impact of a human coagulation factor in nonclinical species during drug development and provides insights into potential implications in human if a similar ADA occurs. FXaI16L-induced ADA response was characterized in monkey, mouse, rat, and dog in different studies, and ADA effects on pharmacokinetic and/or pharmacodynamics of FXaI16L were further examined in ADA-negative and ADA-positive animals. After repeated administrations, FXaI16L elicited a dose and exposure day-dependent ADA response which ranged from no response to a transient or persistent response. Increase in exposure day and increase in dose generally enhanced ADA incidence except for a decrease in ADA incidence was observed in monkeys after repeated high-dose administrations. The observable ADA impact on pharmacokinetics was only found in some ADA+ animals and included decrease in clearance and increase in systemic exposure but no increase in half-life. In addition, no or limited effect on pharmacodynamics by ADA was observed. The earliest ADA response was observed after three exposure days, marked elevation of drug exposure was observed in some animals at log titer > 2.0, and the highest antibody titer excited was about 4 (Log10) in all species. A correlation between ADA induction and accumulative exposure after various repeat treatments in different species was found for FXaI16L. In addition, potential immunogenicity risk and mitigation of ADA in clinics are discussed.

Alemtuzumab depletion failure can occur in multiple sclerosis.

Alemtuzumab is a lymphocyte-depleting antibody and one of the most effective treatments for relapsing multiple sclerosis. However, it also causes loss of immune-tolerance leading to secondary autoimmunity and marked anti-drug antibody responses. Although these anti-drug responses have been reported to be of no significance, we hypothesized that they will affect the depleting capacity and treatment response in some individuals. This was found following analysis of the regulatory submission of the pivotal phase III trials, which was obtained from the European Medicines Agency. At the population level there was lack of influence of 'ever-positive' alemtuzumab-specific antibody responses on lymphocyte depletion, clinical efficacy and adverse effects during the 2-year trial. This was not surprising as no one before the first infusion, and only 0·6% of people before the second-infusion, had pre-infusion, neutralizing antibodies (NAbs). However, at the individual level, NAbs led to poor lymphocyte depletion. Importantly, it was evident that 31% of people had NAbs and 75% had binding antibodies at the end of treatment-cycle 2, which suggests that problems may occur in people requiring additional alemtuzumab cycles. In addition, we also identified individuals, following 'post-marketing' alemtuzumab use, whose lymphocyte level was never effectively depleted after the first infusion cycle. Hence, although alemtuzumab depletes lymphocytes in most individuals, some people fail to deplete/deplete poorly, probably due to biological-response variation and NAbs, and this may lead to treatment failure. Monitoring depletion following infusion and assessment of the neutralizing response before re-infusion may help inform the decision to retreat or switch therapy to limit treatment failure.

Abstract 595: Next-generation engineered toxin bodies: CD38, PD-L1 and HER2 targeted ETBs

Abstract Molecular Templates is developing engineered toxin bodies (ETBs), potent recombinant immunotoxins that combine the specificity of an antibody fragment with the powerful direct cytotoxicity of the Shiga-like toxin A subunit to specifically kill target expressing cells. Once delivered to appropriate cells, the Shiga toxin A subunit inhibits protein synthesis and promotes apoptosis of tumor cells. Bacterial or plant derived toxin moieties have the potential to induce an immune response, commonly limiting repeat dosing of immunotoxins. Our next-generation ETB scaffold has been modified to overcome this limitation through genetic engineering to systematically and comprehensively reduce B and CD4+T cell epitopes. Molecular Templates has developed a proprietary epitope class switching technology designed to both reduce the anti-drug response and promote the anti-tumor response by replacing naturally occurring CD4+ T cell epitopes with CD8+ T cell epitopes. We have found that a combination of surface remodeling and epitope class switching combined in one protein results in powerful reductions in the anti-drug response against ETBs after repeat administration. This de-immunized next-generation scaffold retains the potency and specificity of the unmodified ETB. Additionally, the engineering of CD8+ T cell epitopes on the ETB scaffold can allow for foreign antigen presentation in complex with MHC class I on the tumor cell surface. This antigen presentation may recruit activated cytotoxic T lymphocytes (CTLs) to the tumor microenvironment thereby adding an additional mechanism of action to the direct cell kill activity of the ETBs. The second generation ETB scaffold has been combined with multiple target binding domains, including scFvs that target CD38, PD-L1 and HER2. CD38 is a validated target for multiple myeloma, and our CD38 targeted ETB has shown potent activity in vitro and in vivo. Pre-clinical studies have shown effective combination of the CD38 targeted ETB with standard of care agents such as IMiDs. Additionally, we have developed a PD-L1 targeted ETB that can be used against various PD-L1 expressing malignancies including Hodgkin's lymphoma and melanoma. HER2 is a well characterized target in breast cancer that persists in many patients after HER2 targeted treatment relapse; a novel MOA to this target may impart additional clinical benefit. The potency, reduced immunogenicity, unique mechanism of action and immune modulating activities of ETBs in this second generation scaffold allows for these agents to fit in well in a refractory setting as well as in combination with other agents in the growing field of immuno-oncology. Citation Format: Sangeetha Rajagopalan, Garrett L. Robinson, Brigitte Brieschke, Jennifer Erdman, Jane Neill, Jack P. Higgins, Erin K. Willert. Next-generation engineered toxin bodies: CD38, PD-L1 and HER2 targeted ETBs. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 595.

Manifestations of Antidrug Antibodies Response: Hypersensitivity and Infusion Reactions.

Immunogenicity of biological agents leads to the development of antidrug antibodies (ADA) and it may be associated to hypersensitivity reactions. Immediate infusion reactions occur during or within 1 h after infusion, and their clinical manifestations vary considerably, ranging from mild to severe and life-threatening. Recent studies show that different mechanisms sustain hypersensitivity reactions toward biologics, and the application of novel methods for detecting ADA has demonstrated the involvement of specific IgE isotypes. Considering the severity of the reactions, it is important for clinicians to recognize their symptoms, to know their pathophysiological mechanisms, and to take risk assessment and prophylactic procedures. This review summarizes the clinical manifestations of antibody and nonantibody-mediated reactions as well as the humoral and cellular mechanisms of antidrug responses. Last, the management of patients at risk is discussed. The definition of diagnostic and prophylactic strategies represents an unavoidable need in the management of potentially reactive patients to improve the safety profile of biologics.

The histone acetylranseferase hMOF acetylates Nrf2 and regulates anti‐drug responses in human non‐small cell lung cancer

The histone acetyltransferase MOF is a member of the MYST family. In mammals, MOF plays critical roles by acetylating histone H4 at K16 and non-histone substrates such as p53. Here we have investigated the role of MOF in human lung cancer and possible new substrates of hMOF. Samples of human non-small cell lung cancer (NSCLC) were used to correlate MOF with clinicopathological parameters and NF-E2-related factor 2 (Nrf2) downstream genes. 293T-cells were used to study interactions between MOF and Nrf2, and acetylation of Nrf2 by MOF. Mouse embryonic fibroblast and A549 cells were utilized to assess involvement of MOF in antioxidative and anti-drug responses. A549 cells were used to analysis the role of MOF in anti-drug response in vitro and in vivo. hMOF was overexpressed in human NSCLC tissues and was associated with large tumour size, advanced disease stage and metastasis, and with poor prognosis. hMOF levels were positively correlated with Nrf2-downstream genes. MOF/hMOF physically interacted with and acetylated Nrf2 at Lys(588) . MOF-mediated acetylation increased nuclear retention of Nrf2 and transcription of its downstream genes. Importantly, MOF/hMOF was essential for anti-oxidative and anti-drug responses in vitro and regulated tumour growth and drug resistance in vivo in an Nrf2-dependent manner. hMOF was overexpressed in human NSCLC and was a predictor of poor survival. hMOF-mediated Nrf2 acetylation and nuclear retention are essential for anti-oxidative and anti-drug responses. hMOF may provide a therapeutic target for the treatment of NSCLC.

Investigations of survivin: the past, present and future

Survivin is a member of the inhibitors-of-apoptosis protein (IAPs) family. It promotes cell survival through interference with multiple cell cycle-related proteins such as INCENP and Aurora B kinase. Survivin also inhibits cell death through interference with both caspase-dependent and -independent cell apoptosis. Interestingly, recent evidence suggests that survivin may also play a role in the regulation of cancer cell autophagy. At the clinical level, studies on clinical specimens have shown that survivin expression is up-regulated in various human cancers and its up-regulation is associated with tumour resistance to both chemotherapy and radiation therapy. On the basis of these findings, survivin has been proposed as an attractive target for new anti-cancer interventions. However, despite the role that survivin plays in cancer cell survival and anti-drug response, the development of survivin inhibitors is relatively slow as compared to other therapeutic inhibitors for cancer treatment. In this review, the relationships between survivin expression and the causation of drug resistance in cancers are re-addressed. This review also summarizes the recent development of survivin inhibitors for clinical usage.

Targeting Hsp90 with small molecule inhibitors induces the over-expression of the anti-apoptotic molecule, survivin, in human A549, HONE-1 and HT-29 cancer cells

BackgroundSurvivin is a dual functioning protein. It inhibits the apoptosis of cancer cells by inhibiting caspases, and also promotes cancer cell growth by stabilizing microtubules during mitosis. Since the molecular chaperone Hsp90 binds and stabilizes survivin, it is widely believed that down-regulation of survivin is one of the important therapeutic functions of Hsp90 inhibitors such as the phase III clinically trialed compound 17-AAG. However, Hsp90 interferes with a number of molecules that up-regulate the intracellular level of survivin, raising the question that clinical use of Hsp90 inhibitors may indirectly induce survivin expression and subsequently enhance cancer anti-drug responses. The purpose of this study is to determine whether targeting Hsp90 can alter survivin expression differently in different cancer cell lines and to explore possible mechanisms that cause the alteration in survivin expression.ResultsHere, we demonstrated that Hsp90 inhibitors, geldanamycin and 17-AAG, induced the over-expression of survivin in three different human cancer cell lines as shown by Western blotting. Increased survivin mRNA transcripts were observed in 17-AAG and geldanamycin-treated HT-29 and HONE-1 cancer cells. Interestingly, real-time PCR and translation inhibition studies revealed that survivin was over-expressed partially through the up-regulation of protein translation instead of gene transcription in A549 cancer cells. In addition, 17-AAG-treated A549, HONE-1 and HT-29 cells showed reduced proteasomal activity while inhibition of 26S proteasome activity further increased the amount of survivin protein in cells. At the functional level, down-regulation of survivin by siRNA further increased the drug sensitivity to 17-AAG in the tested cancer cell lines.ConclusionsWe showed for the first time that down-regulation of survivin is not a definite therapeutic function of Hsp90 inhibitors. Instead, targeting Hsp90 with small molecule inhibitors will induce the over-expression of survivin in certain cancer cell lines and subsequently enhances the ability of cell survival in drug-treated situations. The current study suggests that dual inhibition of Hsp90 and survivin may be warranted.

Immunogénicité de protéines d’intérêt thérapeutique

Unwanted immunogenicity, i.e., the developpement by patients of anti-drug antibodies is a significant problem with biologicals therapeutic reagents and can compromise clinical response. Over 20 antibodies currently on the market and over 100 drug candidates are currently in clinical trials; all therapeutic antibodies are showing some level of immunogenicity, and although it has been reduced with the advent of antibodies including human sequences, or even humanised antibodies, this concern will not be totally eradicated. Whereas the actual anti-drug response can only be addressed during clinical development or post-marketing, the industry and the regulatory instances are facing a challenge to develop accurate procedures for the assessment of immunogenicity related to antibody therapeutics, addressing both the likelihood and the severity of the drug-related immunogenicity. This review will discuss the multiple factors that can contribute to a potential immunogenicity of protein therapeutics patient/disease related, as well as related to the drug itself, and the strategies to identify anti-drug antibodies, both in clinical and non clinical assays.

Anti-Drug Antibody Responses to Lumiliximab Are Not Detected in Relapsed Refractory CLL Patients Treated with Lumiliximab in Combination with FCR in a Phase 1/II Study

Lumiliximab, an anti-CD23 monoclonal antibody that binds specifically to human CD23, is under investigation for the treatment of patients with relapsed refractory chronic lymphocytic leukemia (CLL). The clinical safety monitoring for Lumiliximab includes testing for the presence of anti-drug antibodies (ADAs) in patient serum samples. In order to determine if patients develop an ADA response to Lumiliximab, serum samples were collected at specific time points throughout a Phase I/II study (152–30) of Lumiliximab in combination with FCR in relapsed CLL. Patient sera from this Phase I/II study were tested in a standard bridging ELISA in which samples were added to a drug coated plate and ADAs were detected using conjugated drug and colorimetric detection. Validation results showed the sensitivity of the assay was 1000ng/ml using a purified polyclonal rhesus anti-Lumiliximab antibody in undiluted human sera. ADAs were not detected in any patients that had evaluable post treatment (Week 13, Week 25, Week 41, and Month 12 post first dose) serum samples in the 152–30 study. Over the last few years, there has been an effort in the immunogenicity field to develop ADA assays with increased sensitivity and to adapt formats that, in general, have increased drug tolerance, i.e. reduction of drug interference in the assay. Therefore, a new ADA assay was developed using Meso Scale Discovery (MSD) technology that has become the reliable and current method for detection of ADA responses and incorporates an acid dissociation step that has been shown in literature to improve drug tolerance. Using the new method the same evaluable samples from the phase I/II study were tested to determine if any previously unseen ADA responses could be detected. The samples were incubated with biotinylated and ruthenylated drug, added to a streptavidin coated plate and ADAs were detected using electrochemiluminescence. Validation results show that the newly developed MSD assay has sensitivity of 250ng/ml using a purified polyclonal chicken anti-Lumiliximab antibody in undiluted human sera. All the samples tested in the MSD assay (n=32) were negative despite greater sensitivity than the ELISA method and an acceptable drug tolerance at the time points samples were drawn. PK concentrations at the specified ADA collection times ranged between 0μg/ml and 114μg/ml whereas drug tolerance ranges from 1μg/ml to 200μg/ml depending on ADA concentrations. Drug interference is a major factor in not being able to detect antidrug responses, yet several of the corresponding post dose PK time points showed low to no drug levels and yet no anti-drug antibodies were detected. These data imply that Lumiliximab when administered in combination with FCR has a low immunogenicity rate.