Symposia and Workshops

Session Chairs: Raluca Kubaszky, Charles River, Laval, QC, Canada; and Brian J. Mulhern, SciLucent, LLC, Herndon, VA

Gene and cell therapy products are making strides in drug development, demonstrated by a rapidly growing number of such products in development and the recent approvals of products such as Kymriah®, Yescarta®, and Zolgensma®. Given the advancements and growing enthusiasm for these products, this symposium is focused on discussion of the latest hot topics and developing trends in the nonclinical safety assessment of gene and cell therapies. First, a case study of a genetically-modified cell therapy product will be presented to highlight strategies for the design and execution of a nonclinical program. Next, a framework for evaluating biodistribution and safety pharmacology of CRISPR/Cas9 gene-edited products will be discussed. Third, lessons learned from the development and approval of engineered T-cell therapies will be reviewed to shed light on the safe use and toxicity management of these products. Lastly, unique preclinical studies required to support clinical development of gene and cell therapies will be discussed.

1-1

Cell and gene therapy products are on the cutting edge of drug development, providing ways to deliver drugs, alter biologic/genetic processes with long-lasting effects, and/or regenerate function. They also provide unique challenges to preclinical safety programs to support first-in-human trials, including evaluating the effects of ex vivo manipulation and demonstrating biological activity of products of human origin in animal models. To successfully develop and execute a program that meets regulatory expectations, consideration of the specific cell product attributes and intended indication is mandatory. Therefore, cell therapy preclinical safety programs require customized approaches with specialized animal models and/or hybrid efficacy/safety models and sensitive biomarkers of detection for biodistribution, engraftment, and/or expression. In this presentation, a genetically modified cell therapy product will be highlighted to discuss effective strategies for the design and execution of preclinical safety studies for these cellular therapeutics.

1-2

Highly specific genome editing promises to resolve the root cause of many genetic diseases. While precise editing can influence disease outcomes, other biological consequences may result. Effects of on-target editing can vary depending on cell type. For in vivo editing approaches, distribution of editing cargo will influence which cell types are subject to editing. Subsequent distribution of the edited cells can verify durable effects in specific tissue compartments. Experience with in vivo–edited transthyretin (TTR) will be used as an illustrative example. Outcomes of biodistribution studies can help formulate safety assessments and follow-up studies to evaluate the impact of editing. Toxicology studies can also be informed by biodistribution results indicating editing in unintended cell types. Preclinical model selection may present novel challenges for properly evaluating a CRISPR/Cas9 edited product. Model selection can aid in evaluating phenotypes associated with target biology, which can also support true safety pharmacology assessments. A framework for considering biodistribution and safety pharmacology for CRISPR/Cas9 gene-edited products is proposed in this discussion.

1-3

The successful marketing approval of two CD19 CAR T-cell therapeutics has established an important new option for treating otherwise intractable cancers. These products also represent the successful convergence of molecular, genetic, and cellular engineering efforts to reshape cellular functionality. This synthetic biology evolution continues with both the extension of the current technologies to novel cell types and the expansion of novel functionality into the T cell, with the hope of broadening and deepening responses in specific cancers. This talk will review the design of engineered TCR T cells and chimeric antigen receptor (CAR) T cells, with a focus on engineering and manufacturing aspects designed to confer specific pharmacology on the T cell. Lessons learned from translational biology and correlative analyses provide insight into process-, product-, and patient-specific factors that underlie both safety and efficacy. An understanding of key class-specific toxicities is emerging, which can inform the safe use and toxicity management of these therapies.

1-4

Gene therapy for treatment of cancer diseases has grown in importance over the last decade and has entered clinical application. This is reflected by the fact that more than 60% of all gene therapy trials are aiming at treatment of cancer and that their number is still increasing worldwide. In this regard, knowledge on safety, biodistribution, and clearance of vectors for gene therapy is essential. In cancer gene therapy, viral and nonviral vectors are well-established technologies for effective introduction of the respective transgenes. Particularly, nonviral vectors are of increasing attractiveness. This is due to new developments in vector design (minicircles, miniplasmids, minimalistic vectors, etc.) to improve vector transduction and stability as well as transgene expression, and transfer technologies (e.g., physical, liposomal, lipoplex technologies), which aid gene transfer to the tumor cells. Currently, 18% of gene therapy trials employ nonviral vectors. In vivostudies with cell line–derived xenograft (CDX) and patient-derived xenograft (PDX) models, and more importantly clinical gene therapy trials, generate important data on pharmacodynamics, biodistribution, clearance, and stability of nonviral vectors, as important parameters for efficient and safe application of such vectors for future clinical applications. Such data will have an impact on design and performance of future clinical gene therapy trials to better control vector use and to improve therapeutic efficacy in patients. Cell therapies are mainly based on human stem cells, for which the loss of tumorigenicity and teratogenicity has to be shown. Tumorigenicity and teratogenicity assays will be discussed, and case studies involving cell therapy products will be presented.

Session Chairs: Michael Santostefano, Merck & Co., Inc., Boston, MA; and
Debie Hoivik, Akebia Therapeutics, Cambridge, MA

Drug development has been changing considerably over the past several years. There has been an ever-increasing shift from the approval of primarily new chemical entities to protein therapeutics, such as monoclonal antibodies and fusion proteins, as well as entities at the interface of chemicals and biologics (e.g., oligonucleotides and protein-drug antibodies). In parallel to the emergence of these more novel agents, there has been considerable development and application of emerging digital technologies, such as digital pathology, tissue image analysis, and machine learning. This session will provide a snapshot of considerations regarding the development and applicability of some of these modalities and technologies to drug discovery and development. Specifically, this session will discuss development aspects of oligonucleotides and small molecules targeting RNA splicing machinery and how machine learning and artificial intelligence can utilize general pharmacology data to predict therapeutic liabilities or aid in the discovery of safer candidates. Nucleic acid–based and tumor-targeted therapeutic programs offer a unique opportunity to explore consistencies and variances in the pharmacokinetic and toxicity profile. From a technology perspective, digital pathology will display how these tools and artificial intelligence approaches are being applied, and advantages and disadvantages will be demonstrated via case examples. While these agents and technologies are not entirely new, their development and utilization are helping to more quickly bring safer and more directed medicines to patients. Science is changing rapidly, and this session will help to ensure therapeutic safety scientists are prepared for drug discovery and development in the 21st century.

2-1

Decades in the making, nucleic acid–based therapeutics are now beginning to deliver on their considerable promise to become a third significant modality class, alongside small molecule and biologic classes, with applications across many therapeutics areas. Several antisense oligonucleotide (oligos) drugs have already received regulatory approval and are on the market, and the first RNAi US FDA approval was granted in 2018. In addition, numerous mRNA and CRISPR therapeutic programs have entered clinical stages of development. Despite the commonality of being nucleic acid based, the safety profiles can differ, and broad generalizations of toxicity-related findings should be carefully considered. This presentation will provide an overview of some of the preclinical safety considerations from a development and regulatory perspective, with specific focus on mechanisms of toxicity and the pharmacology and toxicity studies that are recommended for inclusion in Investigational New Drug (IND) applications. Topics to be discussed include the commonalities and differences among nucleic acid–based therapeutics; current approaches to address specific and nonspecific toxic effects in nonclinical safety studies; and nonclinical regulatory considerations from FIH to carcinogenicity studies.

2-2

RNA as a drug target offers opportunities to modulate numerous cellular processes, including those linked to the so-called undruggable protein targets. Targeting RNA with small molecules may drastically expand the toolbox to design safe and efficacious medicine for almost any disease. Currently, approved protein-targeted drugs interact with fewer than 700 gene products, and an estimated 10%–15% of proteins are thought to be disease related. Altering the natural process of pre-mRNA splicing can change the formation of splicing products and influence the level of the corresponding protein products. Risdiplam, a small molecule RNA splice modifier, has been developed to interact with the RNA/protein splicing machinery to shift RNA splicing of the survival motor neuron 2 (SMN2) gene to include exon 7 and to generate a functional protein. This is expected to be disease modifying as the SMN1 gene is mutated and dysfunctional in patients with the rare spinal muscular atrophy (SMA) disease. The journey to arrive at this clinically efficacious small molecule involved surpassing safety and selectivity hurdles, which are unique in the combination of these challenges in drug development.

2-3

Patients, product manufacturers, and regulatory authorities share a demand for deep characterization of potential health risks associated with chemical or pharmaceutical products. Over the past several years, the classical toxicity testing of substances has therefore been increasingly complemented by in vitro and in silico methods. Machine learning, particularly deep learning, has been recognized as a key technology for artificial intelligence applications and digitalization. We will describe a deep learning (DL) framework implemented and broadly validated in support of drug research projects. Typical applications are the early identification of drug liabilities and contribution to understanding of liabilities, such as safety pharmacology, hepatobiliary metabolic toxicity, off-target mediated clastogenicity, and preclinical photo-safety. In our experience, DL often outperforms benchmark methods. Particularly, “Multitask”-DNNs can be tweaked to describe multiple endpoints simultaneously in a single framework, further improving the performance of individual models. Apparently this approach capitalizes on hidden trends or correlation even between orthogonal ADME-toxicology assays or endpoints. This feature can be particularly rewarding when working with small toxicology datasets. Our models are managed and deployed in the Google TensorFlow environment, which also enables broad application to novel research compounds.

2-4

Digital computational pathology, a high-velocity, rapidly progressing field disrupting classical diagnostics, has significant ramifications for clinical diagnosis and drug safety assessment. Whole-slide image (WSI)–based digital pathology systems (DPS) are basically composed of commercially available technology that enables a traditionally prepared pathology tissue section mounted on a glass slide to be optically scanned through digital image capture. This permits a user to then pan, magnify, and view the slide image from a database on a digital display monitor, analogous to conventional microscopy. Capabilities permit traditional pathology assessments to be accomplished using readily recalled image data from related studies for side-by-side comparisons. Additionally, remote telepathology consultation and image annotation can be executed in real time with colleagues having network access at other company sites. DPS are forecast to grow at a compound annual rate of 12.1% from 2016 to 2021. Drivers include rising cancer prevalence and increasing demand for telepathology consultations, as well as utilization for drug discovery and companion diagnostics. In addition to database organization and retrieval of histopathology images, the greatest impacts are anticipated to be due to image analysis and pathology informatics now merging into pharmaceutical research, computer-assisted clinical diagnostics, and education. While regulatory policy continues to prescribe conventional microscopy for traditional assessment of test article compound–exposed tissues from in vivo preclinical studies, this fact is likely to be impacted in the future, as examination of WSI moves further into the realm of patient primary clinical diagnosis. Notwithstanding current regulatory policy, discovery and nonregulatory studies benefit from DPS capacity to enable greater objectivity through more reproducible, quantitative computational data that can be generated from digitized images. Furthermore, there is a dynamic spectrum of data management and analysis capabilities for WSI providing the potential for image analysis automation and artificial intelligence applications in computational image data development that move beyond visual inspection of tissue alone. The objective of the presentation will be to indicate how WSI analysis technology is being validated for clinical and preclinical diagnosis, using example approaches and data that can be generated to inform discovery science.

2-5

Digital pathology and image analysis not only have impacted diagnostic pathology and academic research but also are commonly utilized in drug development research. This encompasses aspects such as target expression analysis via immunohistochemistry in human and animal tissues, analysis of early efficacy studies and toxicology studies, and digital pathology in the peer-review process. Most recent advances in this field have been driven by the application of machine learning and deep learning. These fields within artificial intelligence (AI) have been demonstrated to be powerful tools when it comes to histopathology. We will give an overview of cases where AI-powered tools can be useful in driving timely decision-making by generating robust and reproducible data and improving workflow efficiency. Using these examples, we will outline important points to consider regarding implementation of digital pathology, including aspects of collaboration, slide evaluation, image analysis, and AI. The presentation will be divided in two parts. Part I will focus on introduction of the topic, basic concepts, and the current state of digital peer review. Part II will focus on examples of machine learning tools and their use cases in pathology.

Session Chairs: Armaghan Emami, US FDA/CDER, Silver Spring, MD; and
Denis Roy, SciLucent, LLC, Herndon, VA

The safety assessment of excipients in drug products can be challenging when evaluating new formulations, new routes of administration, new indications (acute versus chronic), or new populations. Even with the publication of guidance documents on nonclinical safety evaluation of pharmaceutical excipients, there are still numerous examples where drug developers meet unexpected challenges or obstacles, often attributable to improper characterization of the risks and/or inadequate understanding of the regulatory requirements surrounding excipients. This Workshop provides some key concepts and practical approaches in assessing the safety of excipients from both a systemic and a local toxicity concern. The Workshop will highlight challenges and limitations of the Center for Drug Evaluation and Research (CDER) Inactive Ingredients Database (IID), leveraging data in Master Files, novel routes of administration (such as transdermal or topical delivery systems), and pediatric formulations. It will specifically include safety assessment of flavors in drug products and the limitations of various generally recognized as safe (GRAS) designations commonly submitted to support the safety of excipients contained in Investigational New Drug products. This Workshop will also present industry and US FDA/CDER review division case studies that illustrate current and applicable challenges and how to effectively design a robust approach to the safety of excipients in new therapeutic formulations.

3-1

This introductory presentation will provide an overview of the regulatory expectations for drug product excipient safety assessment, including a discussion of the existing guidance document recommendations, limitations of the CDER IID and GRAS designations, and challenges faced by regulators when sponsors propose the use of new or novel excipients. This presentation will discuss the regulatory considerations in the qualification requirements for “new” excipients as defined by the existing guidance documents. The presentation will illustrate these challenges by drawing upon examples of challenges faced by both the Agency and sponsors of drug products during development, including reliance on the CDER Inactive Ingredients Database, GRAS designations, justification of safety for flavoring agents, and attempt to leverage data in Drug Master Files (DMFs).

3-2

This presentation will provide an industry perspective on the challenges of meeting regulatory agency expectations to support the use of excipients in therapeutic products. The key aspects of excipient safety considerations and requirements will be integrated with real-life examples of development challenges, including novel excipients, reformulations, leveraging GRAS-listed excipients, use of novel or known excipients in delivery systems, and/or use in special routes of administration. Common mistakes/pitfalls companies encounter in leveraging safety information on excipients, including the use of regulatory precedents (e.g., approved products), literature information, and/or dietary/food safety databases (acceptable daily intakes [ADI], permissible daily exposures [PDE], and GRAS), will also be presented, along with practical approaches to avoid product development delays or regulatory issues.

3-3

As the use of dermal patches as a drug delivery system increases, toxicology assessments of drugs administered via this route play an important role in evaluating the potential local adverse effects to patients. This presentation provides a practical illustration of the regulatory requirements associated with development of drugs intended for this route and case studies that illustrate associated challenges as well as approaches to addressing complex scenarios when qualifying the safety of excipients. The presentation will also include a discussion on how to evaluate safety margins based on local and systemic toxicity of excipients.

3-4

Since 1960, the Flavor and Extract Manufacturers Association (FEMA) has supported an independent, continuous program of safety evaluation of flavoring substances. Under a US FDA contract, FEMA compiled safety data on all existing flavoring substances and organized these data into 69 monographs on congeneric groups exhibiting similar biochemical and toxicologic properties. In 1995, US FDA proposed a novel global flavor safety evaluation program to the Joint FAO/WHO Expert Committee on Food Additives (JECFA), which subsequently adopted the procedure in 1996. Flavors were first classified according to one of three classes of relative toxicity using the Cramer et al. (1978) Decision Tree (CDT). The intake of a substance is compared with the Threshold of Toxicological Concern (TTC) determined for its CDT class. If intake is below the TTC, a substance is concluded to be safe; if above the TTC, the substance is evaluated based on its data taken in the context of data for the congeneric group. To date, approximately 3,000 substances have been successfully evaluated and added to a global positive list of flavoring agents. An updated, expanded version of the CDT and new TTC values will provide an additional tool in safety evaluation of all chemical substances with low-exposure scenarios.

3-5

NMEs often behave differently in different populations, but what about excipients? This talk will focus on some of the considerations that should go into drug product design where special populations are involved. Pediatric and geriatric patients will be discussed, as well as other populations that may be adversely affected by excipients such as alcohols, DMSO, cyclodextrins, and propylene glycols. We’ll also delve into different routes of administration and how excipient use can and should be changed to accommodate these special use situations.

3-6

Under the Pediatric Research Equity Act (PREA) (21 U.S.C. 355c), all applications for new active ingredients (which include new salts and new fixed combinations), new indications, new dosage forms, new dosing regimens, or new routes of administration are required to contain an assessment of the safety and effectiveness of the product for the claimed indication(s) in pediatric patients unless this requirement is waived, deferred, or inapplicable. Under the Food and Drug Administration Safety and Innovation Act (FDASIA), a pediatric study plan (PSP) must be submitted within 60 days of an End-of-Phase-2 (EOP2) meeting that outlines the pediatric studies planned or a request for a deferral, partial waiver, or waiver of studies. In addition, the PSP should specifically provide a justification for why nonclinical juvenile animal studies are or are not needed to support the pediatric drug development program, and this should include an assessment of the drug product excipients. This presentation will discuss the regulatory expectations for the nonclinical information needed to justify the safety of excipients intended for pediatric patients and provide examples of appropriate strategies to test these compounds in juvenile animal studies to support dosing in this patient population.

Session Chairs: Vincent Murphy, STILLMEADOW, Inc., Sugar Land, TX; and
Mellessa M. Miller, University of Memphis, Bartlett, TN

Every day, we use water for drinking, bathing, cooking, and cleaning. We trust that those who oversee our water supply maintain its quality and that it is safe for our families’ use. Unfortunately, quality can be compromised when water supplies are contaminated. Contamination has resulted from changes in water supply, delivery, and treatment procedures, as well as from natural disasters like floods and hurricanes and spills from industrial and mining sites. Of interest to toxicologists is contamination from heavy metals and industrial or agricultural chemicals. This Symposium will highlight some of the recent contamination incidents—including the lead contamination of the Flint, Michigan, water supply and industrial chemical contamination at Camp Lejeune in North Carolina and the Fox River in Wisconsin—and it will explore how the water supply was compromised and what health concerns resulted from the contamination. Additionally, speakers will discuss the remedies to improve the water quality and how to avoid a future occurrence. Research into evaluating the exposure and risk to the population will be presented.

4-1

Ever since a citizen science collaboration between Flint, Michigan, residents and the Virginia Polytechnic Institute and State University (Virginia Tech) research team exposed citywide water lead contamination in August 2015, there has been widespread concern regarding consumer exposure to lead during the Flint Water Crisis (FWC). The FWC was triggered when the City of Flint switched its water supply from Lake Huron to the Flint River in April 2014 without implementing federally mandated corrosion control treatment, causing a spike in lead levels in Flint’s tap water. It is acknowledged that water lead sampling approaches used violated federal law, rendering the available official data useless. The nearly complete lack of data on lead levels in Flint’s tap water and associated uncertainties with human exposure from April 2014 to August 2015 have ultimately led to intense speculation, proxy research, and controversy. Proxy research using blood lead measurements by Flint pediatricians, CDC, and others has tied the source water switch to a doubling of the percentage of children less than six years old with elevated blood lead (EBL; >= 5ug/dL) among other illnesses and deaths, in the now internationally acknowledged case of environmental injustice. Depending on the news source, one may gather that all Flint’s children were “poisoned” or that the persistent spotlight on FWC was tantamount to “toxicohistrionics.” Understanding what occurred during the FWC, therefore, remains of great scientific, social, political, and legal interest. This presentation will highlight a novel approach that utilizes routine monthly lead in sewage sludge (or biosolids) monitoring data to examine lead release to sewage via potable water. We successfully link lead in biosolids to water lead levels pre-, during, and post-FWC, which is consistent with the changes in children’s blood lead. The analyses confirmed expectations that other lead sources (i.e., paint, dust) were probably major contributors to blood lead at times other than summer 2014, even though higher lead in water persisted through at least October 2015, when public health interventions, including provisions for bottled water and lead filters, were implemented. Furthermore, we use an age-specific biokinetic model to mimic low, typical, and extreme lead exposure scenarios seen pre-, during, and post-FWC to estimate adverse health outcomes in young children and pregnant women. Finally, we use data and findings from historical events like the USA Today building miscarriage cluster (1987–1989) and Washington, DC, lead in drinking water crisis (2001–2004) to place the FWC in broader historical and public health context.

4-2

Flint, Michigan, switched its public water source in April 2014, increasing exposure to lead and other contaminants. We compared the change in the fertility rate and in health at birth in Flint before and after the water switch to the changes in other cities in Michigan. We found Flint fertility rates decreased by 12% and overall health at birth decreased. This effect on health at birth is a function of two countervailing mechanisms: negative selection of less healthy embryos and fetuses not surviving (raising the average health of survivors) and those that survived being scarred (decreasing average health). We untangle this and find a net of selection scarring effect of 5.4% decrease in birth weight. Due to long-term effects of in utero exposure, these effects are likely lower bounds on the overall effects of this exposure.

4-3

Physiologically based pharmacokinetic (PBPK) models are a well-accepted tool used by regulatory agencies for multiple risk assessment applications. These applications include exposure route extrapolation, exposure scenario extrapolation, and cross-species extrapolation. All of these extrapolation areas are key to performing credible, scientifically sound human health risk assessment, and frequently, all are performed in the process of developing health-protective exposure limits for pollutants in multiple media. In this talk, the utility of PBPK modeling will be illustrated using bromodichloromethane, a drinking water disinfection byproduct, as a case study. This model has multiple risk analysis applications, including multiroute exposure assessment, prediction of toxicity based on internal dose to target tissues, risk analysis for potentially susceptible subpopulations, and prediction of the effect(s) of changes in disinfection scenarios on tissue dosimetry and toxicity.

4-4

The Fox River in northern Wisconsin has accumulated a number of toxins due to waste disposal into the river for over 100 years. This buildup of toxic waste included arsenic, lead, mercury, and polychlorinated biphenyls (PCBs). However, the predominant contaminant in the water that posed the most threat to humans and the ecosystem was found to be PCBs. PCBs biomagnify as they move up the aquatic food chain as larger organisms consume PCB-contaminated food sources, and eventually, this includes humans who ate contaminated fish. Over time, this adversely affected the wildlife and ultimately impacted the human population that lived along the Fox River. Additionally, PCBs are very lipophilic and are stored in fatty tissue, resulting in their bioaccumulation. This fatty tissue accumulation allows for mammals to pass PCBs onto their young in utero and during lactation. This can affect the next generation, especially during critical windows of development that can have lasting implications into adulthood. Clinical studies and research using a developmental PCB-exposure model in rodents will connect the dots of the Fox River contamination and its human impact.

4-5

A US government health agency, the Agency for Toxic Substances and Disease Registry (ATSDR), conducted five epidemiological studies to determine whether volatile organic compound (VOC)–contaminated drinking water exposures at US Marine Corps Base Camp Lejeune, North Carolina, were associated with increased health risks to children and adults. These health studies required information, data, and knowledge of contaminant concentrations in drinking water, at monthly intervals, delivered to family housing, barracks, and other facilities within study areas. Because concentration data were limited or unavailable during much of the contamination time frame (1950s–1985), the historical reconstruction process was used to quantify estimates of monthly mean contaminant-specific concentrations. Examples of results show reconstructed (simulated) tetrachloroethylene (PCE) concentrations reached a maximum monthly average value of 183 micrograms per liter (µg/L) compared with a one-time maximum measured value of 215 µg/L and exceeded the current US EPA maximum contaminant level (MCL) of 5 µg/L during the period of November 1957 to February 1987 at the Tarawa Terrace water treatment plant (WTP). Reconstructed trichloroethylene (TCE) concentrations reached a maximum monthly average value of 783 µg/L compared with a one-time maximum measured value of 1,400 µg/L during the period from August 1953 to December 1984 at the Hadnot Point WTP (also exceeding the current US EPA TCE MCL of 5 µg/L). The epidemiological studies would not have been able to evaluate exposure-response relationships without the monthly mean drinking water concentrations produced by the historical reconstruction process. The ability to evaluate chemical-specific associations and exposure-response trends, rather than simply comparing exposed to unexposed, greatly enhanced the impact of these studies and the evidence they provided.

Session Chairs: Michael Bolt, Pfizer Inc., Cambridge, MA; and
Laurence Whiteley, Pfizer Inc., Cambridge, MA

Gene therapy is the introduction of nucleic acid(s) into a host cell to replace or repair a lost or mutated gene or to introduce something novel altogether. Adeno-associated virus (AAV), a DNA parvovirus, is currently the most commonly used vector for in vivo gene therapies for potential treatment or cure of life-threatening diseases. The purpose of this Symposium is to provide an overview of AAV-based gene therapies and the nonclinical studies that are needed to support clinical trials. This Symposium will include a review of the AAV vector field and the current state of its clinical use, a case study of an ophthalmic AAV gene therapy, and a case study of the biodistribution and toxicity profile of AAV gene therapy following CNS delivery. This Symposium will also discuss the challenges that industry has been encountering for AAV-associated gene therapies. The session will conclude with a panel discussion on the need for harmonization and an ICH guidance on AAV-based gene therapy.

5-1

Since the first proof of concept human application in the early ’90s, the field of gene therapy has now entered a stage of unprecedented revolution for clinical translation and commercialization. Gene therapy can be accomplished through in vivo and ex vivo approaches by gene replacement, for loss-of-function genetic diseases; gene silencing, for gain-of-function genetic disorders; gene editing, for any genetic diseases; and gene addition, for treating acquired diseases. The design and selection of vehicles used to deliver the genetic payload, called vectors, are key elements for successful genetic therapies. The progress of human gene therapy in the past decades has been primarily driven by vector technology platform development. Among the current variety of vectors available for in vivo gene therapy, recombinant adeno-associated virus (rAAV) stands out for its high efficiency, stability, and low immunogenicity/toxicity profiles. AAV is a common benign residential virus that can persist in primate tissues for the lifetime of the host without integration into the host genome and pathological consequences, holding great promise for multiple gene therapy applications. Several recombinant AAV-derived gene therapy drugs have been approved by European and US regulatory authorities for commercialization, including the AAV2-based drug, Luxturna (Spark Therapeutics) for treating a childhood blindness and the AAV9-based Zolgensma (AveXis) for treating spinal muscular atrophy. This presentation will provide an overview of the key principles, history, current challenges, and future directions of human gene therapy, with a focus on rAAV gene therapy, and showcase AAV capsid discovery and engineering to modulate target tissue tropism and biodistribution profiling, therapeutic gene expression cassette design and optimization, rAAV transduction biology, and examples of AAV gene therapy development—from proof of concept preclinical studies to translational studies in large animal models to first-in-human clinical evaluation.

5-2

Retinitis pigmentosa is a form of retinal degeneration usually caused by genetic mutations affecting key functional proteins. We have previously demonstrated efficacy in a mouse model of RLBP1 deficiency with a self-complementary AAV vector carrying the gene for human RLBP1 under control of a short RLBP1 promoter (CPK850). In this communication, we describe the nonclinical safety profile of this construct as well as updated efficacy data in the intended clinical formulation. In Rlbp1^-/- mice dosed at a range of CPK850 levels, a minimum efficacious dose of 3x10⁷ vg in a volume of 1 mL was observed. For safety assessment in these and Rlbp1^+/+ mice, optical coherence tomography (OCT) and histopathological analysis indicated retinal thinning that appeared to be dose dependent for both Rlbp1 genotypes, with no qualitative difference noted between Rlbp1^+/+ and Rlbp1^-/- mice. In a nonhuman primate study, RLBP1 mRNA expression was detected and dose-dependent intraocular inflammation and retinal thinning were observed. Inflammation resolved slowly over time and did not appear to be exacerbated in the presence of anti-AAV antibodies. Biodistribution was evaluated in rats as well as from satellite animals in the nonhuman primate study. The vector was largely detected in ocular tissues as well as at low levels in the optic nerve, superior colliculus, and lateral geniculate nucleus, with limited distribution outside of these tissues. These data suggest that an initial subretinal dose of ~3x10⁷ vg/mL CPK850 could safely be used in clinical trials.

5-3

This presentation will discuss the nonclinical development of an AAV gene therapy for treatment of a neurodegenerative disease. Differences in the biodistribution, transduction efficiency, and toxicity profile in nonhuman primates (NHP) will be described depending on the route of administration, dose level, volume, and animal age. Data suggest that at a physiologically feasible injection volume and regardless of age, transduction efficiency in the spinal cord motor neurons via intrathecal delivery is superior with intra cisterna magna (ICM) injection when compared with lumbar puncture (LP). Transduction efficiency results between NHP studies with AAV-GFP versus our AAV gene therapy on day 21 or 28 post-injection are in agreement, indicating that NHP studies with a reporter gene can be valuable in exploring routes of administration and translating efficacious dose levels from animal disease models. To confirm biodistribution, transduction durability, and potential toxicity in an NHP study with the selected dose and route of administration, testing the clinical candidate with at least one time point beyond that used in studies with AAV-GFP can be considered. Histopathology findings associated with intrathecal delivery of our AAV gene therapy consisted of asymptomatic sensory neuronopathy characterized by neuronal cell body degeneration in dorsal root ganglia and trigeminal ganglia accompanied by axonal degeneration of the dorsal funiculi in the spinal cord and axonal degeneration of peripheral nerves. These histopathologic effects, which were observed independent of the route of administration, emerged between day 14 and 28, plateaued by day 90, and did not progress by day 180 post-injection.

5-4

The EMA and US FDA have published multiple guidance documents that support the development of gene and cell therapies. With the recent approval of several AAV gene therapeutics and many programs in phase 3 development, there is increasing industry experience and interactions with worldwide regulatory authorities. This presentation will provide an overview of regulatory guidelines and results from an industry survey, conducted by the EFPIA Gene Therapy Working Group and IQ Pharmaceutical Consortium, on nonclinical development, and regulatory interactions for AAV gene therapeutics.

5-5

The current guidance for developing gene therapy–based products is consistent with the case-by-case approach introduced in the late ’90s for advancing protein-based pharmaceuticals into the clinic. Although flexible, this approach incorporates the basic toxicological principles that underlie the traditional, standardized preclinical testing paradigm (i.e., similar principles, different practices). The current guidance has served well to advance hundreds of both ex vivo gene-based cellular therapies and in vivo gene therapy–based products, including nonviral vectored therapies, into clinical trials. This includes the various more recent gene-editing modalities. The intended goal is that the preclinical program for each investigational gene therapy product should be individualized with respect to scope, complexity, and design to maximize the predictive value of these studies for clinical safety and therapeutic activity. The unintended goal of developers, albeit necessary, has been to maximize the predictive value of the program acceptable to a regulatory agency. As we continue to advance the novelty of products in this field (e.g., novel vectors, dual vectors, use of more than one promoter and/or multiple transgenes in a single vector), as well as more novel routes of administration and delivery approaches, are we really asking for a standardized preclinical testing paradigm? Designing successful programs is both science based and experience based. Would case studies be more valuable to understand best practices for current products in development and also serve as a reference point for the next generation of products?

Session Chairs: Jessica Hawes, US FDA/CDER, Silver Spring, MD; and
Doris Zane, Intarcia Therapeutics, Inc., Hayward, CA

There has been an increased interest in and activity for the use of peptide therapeutics to treat a variety of human diseases, with innovative strategies for the development of biopharmaceutical pipelines. The number of peptide drugs entering into the market has increased significantly despite inherent challenges of peptide instability and patient-friendly delivery. Disparities in interpretation and application of existing regulatory guidances to innovative synthetic and conjugated products have resulted in challenges for both regulators and sponsors. This Symposium will cover different challenges in peptide therapeutic development, along with considerations of regulatory challenges. The specific topics to be covered will include the following: (1) peptide therapeutic progress and future direction, and approaches to discover, optimize, assess, and deliver combination peptide therapeutics for treatment of diseases; (2) toxicological considerations to deliver peptide drug-device combination products for efficient development and optimal patient benefit and adherence; (3) industry and regulatory perspectives on the regulation of synthetic and conjugated peptide products, including exploration of regulatory classifications, interpretations, and application of existing guidances in determining nonclinical study requirements (ICH M3[R2] versus ICH S6); and (4) discussion of the 2016 HESI working group assessment of genotoxicity testing requirements. Both industry and regulatory professionals will be included as presenters and discussion facilitators.

6-1

There has been surgent interest and activity in the use of peptide therapeutics to treat a variety of human diseases. Approaches to discover, optimize, assess, and deliver combination peptide therapeutics for treatment of metabolic diseases will be presented. Target selection utilizing observations resulting from Roux-en-Y gastric bypass surgery and rigorous preclinical assessment of peptide singleton and combinations will be reviewed. Peptide optimization is conducted to select for highly in vivo selective and potent peptides, and optimized peptide combinations are preclinically assessed to achieve maximal therapeutic benefit. Additionally, various considerations to deliver peptide combinations for efficient development and optimal patient benefit and adherence will be presented.

6-2

Given the inherent physical properties of amino acid polymers and poor stability, delivery of peptide therapeutics to their intended therapeutic targets represents a particularly challenging obstacle. Various toxicological considerations to deliver peptide drug-device combination products for efficient development and optimal patient benefit and adherence will be examined. A brief overview of toxicological expectations, including regulatory requirements, for combination products will be presented. Case study examples and experience with peptide drug-device combination therapeutics, including multipurpose study designs to minimize further animal testing, will be discussed.

6-3

A vast disease target space existing intracellularly is now within the cross-hairs of both academic basic science and pharma/biotech drug discovery efforts. Innovative technologies enabling both chemistry and biology is driving significant progress to advance novel peptide modalities, especially macrocyclic analogs, that have compelling target binding affinities, cell permeabilities, and intracellular proteolytic stabilities. Such properties are critical to optimization and development of a third wave of peptide therapeutics to tackle diseases involving challenging drug targets, such as intracellular protein–protein interactions. Evolving concepts, innovative technologies and progress in both preclinical research and clinical development for this emerging peptide therapeutic modality will be presented.

6-4

Peptide/protein biotherapeutics fall in a gray area between biologics and small molecules when it comes to genotoxicity testing. Perspectives from the Health and Environmental Sciences Institute (HESI) Genetic Toxicology Technical Committee (GTTC) workgroup on synthetic and conjugated peptide therapeutics will be presented. In 2016, the HESI GTTC working group published the article “Genotoxicity Assessment of Peptide/Protein-Related Biotherapeutics: Points to Consider before Testing” in the international journal Mutagenesis. The presentation will discuss the gaps in current regulatory guidelines and provide the workgroup’s opinion on specific case examples.

6-5

Challenges and considerations for peptide therapeutic products will be presented from a regulatory perspective. Disparities in interpretations and application of the existing International Council for Harmonisation (ICH) guidances ICH M3(R2) and ICH S6 to synthetic and conjugated peptide therapeutic products will be explored. Regulatory perspectives and rationale regarding nonclinical studies evaluating genetic toxicology and metabolic characterization will be discussed in further detail with case study examples. The impact of the March 2020 implementation of the Biologics Price Competition and Innovation Act on peptide-related product classifications will be clarified based on the most recent US FDA Q&A draft guidance.

Session Chairs: Pedro L. Del Valle, US FDA/CDER, Silver Spring, MD; and
Bettina Donato, Intertek Group plc, Sacramento, CA

Regional regulations on good laboratory practices (GLP) have been published to ensure the quality and integrity of nonclinical laboratory study data that are intended to support clinical research or marketing permits for human therapeutics. The US FDA promulgated GLP regulations in 1979 and the Organisation for Economic Co-operation and Development (OECD) followed with the publication of GLP Principles in 1981. Many individual nations generated their own unique GLP regulations. Under the Mutual Acceptance of Data (MAD) system, data generated in toxicological testing in one OECD Member country are accepted in all other OECD Member countries. However, several differences in national GLP regulations exist, not all countries are OECD Members, and not all laboratories conduct toxicological testing following US or OECD GLPs, generating hurdles for the acceptance of GLP data in global drug development. The goal of this Workshop is to openly discuss the similarities and differences in GLP regulations and inspections from different countries and regions to facilitate identifying challenges when outsourcing nonclinical GLP testing globally. In this Workshop, the speakers will share their experience and interpretations of GLP regulations and inspections from OECD Member countries (US, Canada, UK, Japan, Korea, and Germany), Brazil (Latin America), Taiwan, and China, by covering (1) GLP roles and responsibilities; (2) testing facility personnel credentials and management responsibilities and training; (3) the quality assurance unit reporting process; (4) SOPs formulation, approval, and enforcement; (5) animal welfare, ISO standards and quality systems, test article characterization, and regulatory inspections; and (6) current challenges in compliance. This Workshop will be of interest to those seeking to understand differences and challenges when considering global outsourcing during drug development.

7-1

This talk will focus on US FDA and contrast with the Medicines and Healthcare Products Regulatory Agency UK (MHRA) regulations and OECD principles. Conduct of nonclinical toxicology studies is becoming more complicated because many test articles have unique characteristics that require analysis at niche laboratories. While multisite study conduct has always been a part of the US FDA GLP regulations, work placed outside the US requires understanding of the similarities and differences between those countries’ regulations. Consideration of the US and UK GLP laws, and comparison with the OECD GLP principles, shows that it is possible for firms to claim compliance of work performed in that country and OECD guidelines. Topic examples include circumstances of when QA audits are shared, when it is acceptable to amend a closed study, and difference in regulatory inspections. The latter follows the US FDA Compliance Program Guidance Manual (CPGM), revised in 2018, and the MHRA Schedule 2 Inspection Procedures. Current challenges include compliance certificates, enforcement of guidance (e.g., MHRA GxP Data Integrity), differing expectations regarding management oversight of risk and enforcement of OECD guidance, and conduct of a non-GLP phase on a GLP study.

7-2

Advances in communication in a rapidly changing world, travel, and technology make it easier to outsource drug development globally. This can be a huge benefit for business and advancement of drug development, but it can pose a risk for regulatory compliance. The OECD GLP Principles are very robust, but their application and interpretation can present certain challenges when outsourcing studies or phases of studies. Careful consideration needs to be taken due to differences between the OECD and US FDA GLP guidelines and differences in interpretation of OECD guidelines by different GLP Monitoring Authorities (MAs) in OECD countries. This presentation aims to provide the audience with a high-level overview of the OECD GLP guidelines (as adopted by the UK) and address the main differences between these, other OECD countries—such as Canada, France, and the Netherlands—and the US FDA GLPs, providing advice on conducting multisite studies with contributions from OECD countries. The most important difference is the OECD defined roles and responsibilities of study personnel and management, which needs to be fully understood before working with sites outside the US. Another critical difference to keep in mind is that OECD MAs have expectations that test item (test article) characterization can be conducted to non-GLP standards. The recent OECD test item characterization document will be referred to as well as updated guidance on data integrity from the MHRA that focuses on the expectation that study directors have detailed knowledge on how electronic data are generated, captured, and archived.

7-3

Taiwan FDA (TFDA) implemented GLP regulations for nonclinical laboratory studies to ensure the quality and commissioned testing activities started in 2016. But TFDA GLP regulations had been put into practice in 2006. Test facilities may apply for accreditation, and once accredited, the test facilities are required to follow the TFDA “Good Laboratory Practice for Nonclinical Laboratory Studies.” There are eight sections in the TFDA GLP regulations: (1) general, (2) organization and personnel, (3) facilities, (4) instruments, (5) operations, (6) test items and references, (7) test plans and execution of tests, and (8) records and reports. The content of the above regulation is comparable to that of international regulations, such as OECD and US FDA GLP. Taiwan is working to become an OECD Member, and TFDA has been working to establish international cooperation relationships with other countries in terms of Good Laboratory Practice for Nonclinical Laboratory Studies. Currently, there are 13 GLP-accredited laboratories in Taiwan. In 2017, TFDA held a “Seminar on Malaysia Pharmaceutical Products GLP Regulations and Principle” to enhance the exchange of information and promote the national cooperation. This talk will elaborate on the current status and future plans of GLP accreditation in TFDA, as well as compare the differences between GLP regulations in Taiwan with UK MHRA regulations, OECD GLP, and US FDA and how those differences can be overcome for global submissions. The challenges and advantages to conducting a GLP-compliant study in Taiwan as part of a multisite study will also be discussed.

7-4

Quality systems in Brazilian test laboratories have been discussed and implemented since 1994 but significantly recognized and accredited after 2000 by national and international organizations. In Brazil, nonclinical studies supporting regulatory submissions, requested by regulatory agencies, such as health, environmental, and agriculture, need to be conducted under GLP rules, by laboratories formally recognized by the National Institute of Metrology, Quality and Technology (INMETRO). In 2011, INMETRO, obtained full adherence to OECD acts related to mutual acceptance of data for chemical evaluation, including nonclinical environmental safety and human health tests for pesticides, their components, and the like. In 2015, the scope was expanded to include veterinary products, feed additives, cosmetics, pharmaceuticals, sanitizers, wood preservatives, and remediators. Thus, Brazil, as a nonmember country of the OECD but with full adherence to its regulations, fulfills duties as recommended by MAD. INMETRO is responsible for inspecting GLP laboratories according to OECD/GLP, as the Brazilian GLP principles are a translation of those of the OECD. There are currently 47 GLP laboratories recognized by INMETRO. Such laboratories keep records of the professional competencies of the personnel involved in carrying on the studies conducted according to OECD Test Guidelines for Chemicals that can be detailed in specific SOPs. Depending on the scope of the laboratory, SOPs for all aspects of GLP must be followed. This presentation will focus on the GLP requirements allowing studies, conducted in Brazil and other Latin American countries, to be accepted by health and environmental regulators adhering to UK MHRA, US FDA, and OECD principles.

7-5

GLP regulations in China were published and implemented to ensure the quality and integrity of nonclinical studies supporting clinical research or marketing approval in China. GLP regulations in China in principle follow the GLP guidelines from the US FDA and the Organisation for Economic Co-operation and Development (OECD). However, there are differences between GLP regulations of China and those of the rest of the world, including OECD and US FDA, such as test facility management (TFM) approval for protocols/reports, additional requirements for animal facility, or requirements for QAU lead. In 2017, Chinese FDA (now National Medical Products Administration, or NMPA) published updated GLP regulations, and major changes were made, including removing the responsibility for TFM approvals, adding requirements on multisite study, computer system, pathology peer review, and sponsor responsibility. New GLPs are more aligned with US FDA and OECD GLP regulations but include requirements from Chinese FDA regarding nonclinical study design, which presents challenges when outsourcing studies in China or studies conducted elsewhere with intention of Chinese submissions. Studies have been rejected for not following Chinese GLP regulations/study guidelines. There are approximately 60 GLP laboratories recognized by NMPA. The regulations in China are evolving, which becomes a challenge for multinational companies as well as biotech companies planning to conduct studies in China. This presentation will discuss interpretation of Chinese GLP regulations; how they differ from UK MHRA, OECD, and US FDA; and how they can be overcome for global submissions.

Session Chairs: Natalie S. Holman, Eli Lilly and Company, Indianapolis, IN; and
Tracy M. Williams, Eli Lilly and Company, Indianapolis, IN

This session is a companion to the successful 2017 ACT Symposium “What to Do When Things Go Wrong,” in which speakers discussed their experiences with test article limitations, publication feedback, formulation challenges, resolution of concerns as regulators, communications as a study director, and unanticipated toxicities on day 1 of a study. This year, early career toxicologists from the pharmaceutical and biotechnology industries as well as the US FDA will present a new set of challenging scenarios they have encountered in their respective roles and how they were resolved. Case studies will include challenges as a new pharmacology/toxicology reviewer, evaluating nontraditional data in GLP toxicology studies, navigating project team dynamics, unexpected pathology findings, and employment instability. Audience engagement and questions are highly encouraged!

8-1

Congratulations—you have worked hard in your career and are now at the point where the problem is your problem! There will always be unforeseen and difficult situations, but now, when something goes wrong, your colleagues look to you for a resolution. Both what you do and how you act can be equally important. The talk will highlight five key principles for dealing with problems: (1) put patients first, (2) identify the information gaps, (3) ask partners for help, (4) determine what can and cannot be fixed, and (5) communicate at the right time. Examples of things going very wrong will be presented, including a fatal tumor on a chronic study, an assay ruined by a disgruntled employee, and a loss of data integrity. In these real-world situations, the five principles were applied at various points to cut through many different opinions and find the path to the best outcome.

8-2

GLP toxicology protocols in support of IND-enabling activities include basic study endpoints guided by an array of ICH Guidelines and health authority recommendations. However, case-by-case needs for novel targets benefit from unique, functional toxicity endpoints or investigative toxicology assays. The requirement to fully assess the safety of novel targets presents not only challenges but also opportunities for the project toxicologist. Clearly defined data collection and evaluation procedures plus diligent oversight for these complex studies may identify critical toxicity signals that would otherwise have been missed and likely enhance the review and acceptance of the toxicology program by health authorities.

8-3

Supporting mature drug products is not as straightforward as one would expect. Toxicological challenges can arise postmarket that require significant time and resources. Two case studies on mature drug products will be presented, along with the risk mitigation plan and regulatory strategy developed to overcome the toxicological challenges. The first case study will describe global expansion of a reformulated drug product with a novel excipient, and the second will describe a postapproval change in drug substance manufacturer that resulted in a change in specification of a potentially genotoxic degradation product.

8-4

Your first job after graduate school seems like the most important career move since deciding to go to school. Who you work for and what you do could have the potential and impact to shape the rest of your career. Working for a biotech start-up company can be either a very exciting or a very risky experience. The speaker will share some experiences of working at a biotech start-up, including (1) accepting a high-risk, high-reward position; (2) dealing with job loss early in one’s career; and (3) reentering the job market with only six months of postgraduate experience.

8-5

After a few years as a toxicologic pathologist, the steepness of the learning curve finally starts to flatten out. You are the program pathologist for a high-priority/high-profile program gearing up for candidate selection. Accurate characterization of target organ toxicities is in your hands, and there is no time to waste. Gulp! Previous nonclinical studies have been performed with this asset, so as you sit down with your cup of coffee to start peer review of a repeat-dose rat study, you think you know what to expect. But wait! Are your eyes deceiving you, or is that a retinal lesion? This is a new and unexpected finding! What does this mean for the program? How can you, the toxicologist, and the rest of the project team work together with urgency to derisk this liability as the program hurtles toward the next stage gate with all eyes on you?

8-6

You finally land your dream job in the federal government but are now faced with dealing with work that is directly related to the well-being and health of many. No other job has prepared you for the work you are about to do. As a brand-new reviewer, how can you make an informed decision and at the same time follow the clues in a sea of data to make a vital decision that could potentially help or harm many human beings? This speaker will walk through interesting and difficult case studies faced as a first-year pharmacology/toxicology reviewer at the US FDA and how to overcome the many obstacles in a brand-new job.