Recommendations for Safety Pharmacology Evaluations of Oligonucleotide-Based Therapeutics

Abstract

This document was prepared by the Safety Pharmacology Subcommittee of the Oligonucleotide Safety Working Group (OSWG), a group of industry and regulatory scientists involved in the development and regulation of therapeutic oligonucleotides. The mission of the Subcommittee was to develop scientific recommendations for the industry regarding the appropriate scope and strategies for safety pharmacology evaluations of oligonucleotides (ONs). These recommendations are the consensus opinion of the Subcommittee and do not necessarily reflect the current expectations of regulatory authorities.

1) Safety pharmacology testing, as described in the International Conference on Harmonisation (ICH) S7 guidance, is as applicable to ONs as it is to small molecule drugs and biotherapeutics.

2) Study design considerations for ONs are similar to those for other classes of drugs. In general, as with other therapeutics, studies should evaluate the drug product administered via the clinical route. Species selection should ideally consider relevance of the model with regard to the endpoints of interest, pharmacological responsiveness, and continuity with the nonclinical development program.

3) Evaluation of potential effects in the core battery (cardiovascular, central nervous, and respiratory systems) is recommended. In general:

a. In vitro human ether-a-go-go-related gene (hERG) testing does not provide any specific value and is not warranted.

b. Emphasis should be placed on in vivo evaluation of cardiovascular function, typically in nonhuman primates (NHPs).

c. Due to the low level of concern, neurologic and respiratory function can be assessed concurrently with cardiovascular safety pharmacology evaluation in NHPs, within repeat-dose toxicity studies, or as stand-alone studies. In the latter case, rodents are most commonly used.

4) Other dedicated safety pharmacology studies, beyond the core battery, may have limited value for ONs. Although ONs can accumulate in the kidney and liver, evaluation of functional changes in these organs, as well as gastrointestinal (GI) and unintended “pro-inflammatory” effects, may be best evaluated during repeat-dose toxicity studies. Broad receptor- or ligand-binding profiling has not historically been informative for most ON subclasses, but may have value for investigative purposes.

Introduction

This document focuses on the Subcommittee's experience and strategies used to assess the safety pharmacology of ONs and is not intended as a primer on the conduct of safety pharmacology studies. The scope encompasses all ONs, regardless of pharmacological mechanism of action, chemical class, or possible mechanisms of toxicity. However, the predominance of information on ONs comes from ONs with a phosphorothioate (PS) or similar highly polyanionic backbone structure and, as such, has influenced the regulatory environment and the recommendations in this document. An attempt is made to distinguish recommendations that are particularly well suited for such ONs, and other ONs that may not exhibit similar class effects. For the most part, the document concentrates on systemic routes of administration, as this is the case for most ONs in development. Strategies for assessing safety pharmacology of ONs that target specific functions of the major organ systems are beyond the scope of this review, although the effect of ONs administered via the respiratory tract and GI tract are discussed briefly under the relevant sections. It is the intent of the Subcommittee to increase awareness and raise points to consider rather than to dictate specific study designs.

Regulations Regarding Safety Pharmacology Testing

Safety pharmacology studies of pharmaceuticals are regulated under several guidances, including ICH M3(R2) (ICH, 2010a), ICH S7A (ICH, 2001), and ICH S7B (ICH, 2005). Additional information is provided for biotechnology-derived products in ICH S6 (ICH, 2011) and for oncology products in ICH S9 (ICH, 2010b). These guidance documents describe the primary objective of safety pharmacology studies as identification of possible acute, undesirable functional effects on the major physiological systems. In general, the core battery of studies, including the assessment of effects on cardiovascular, central nervous, and respiratory systems, should be conducted before human exposure. Although safety pharmacology studies have been traditionally conducted as separate single-dose studies, more recent guidance encourages incorporation of the functional endpoints into general toxicity studies in order to reduce animal use. With regard to species selection, ICH S7A suggests that “consideration should be given to the selection of relevant animal models” (ICH, 2001).

The OSWG Subcommittee agrees in principle with the applicability of these safety pharmacology guidances to ON therapeutics and with the recommended timing for testing relative to clinical development. The Subcommittee considered each physiological system carefully and has made recommendations of appropriate strategies for testing.

Study Design Considerations

In general, the study design considerations for ONs are similar to those for other classes of drugs as outlined in ICH S7A and S7B (ICH, 2001, 2005). The approaches used to select dose levels and the route of administration for ONs should be no different from those used for small molecule drugs. In addition, like small molecules, ONs are typically not immunogenic; therefore, in vivo studies can reutilize animals, which is always preferable when using non-rodents. For those ONs with a PS backbone, parenteral dose levels in NHPs may be limited due to complement activation; therefore, a study design using escalating doses may be more suitable for safety pharmacology studies in NHPs with such ONs, whereas a Latin square design may be acceptable for other types of ONs. The interval between subsequent doses should be considered carefully relative to available information on persistence in tissues and/or duration of pharmacodynamic effect. This type of ascending-dose study design can also serve as a dose range-finding/tolerability study during initial testing in the non-rodent species.

The appropriate group size for safety pharmacology studies is model specific and depends mainly on the sensitivity and variability of the model. Specific commentary about group sizes is included in some of the following subsections, when warranted.

With regard to definition of the test article, the Subcommittee recommends that the drug product be tested. For greatest clinical relevance, conjugated or formulated ONs should be administered as such, not as the naked ON. Under specific circumstances (i.e., when the ON pharmacology is highly human specific), testing of an animal-active analog may be considered to assess possible unwanted effects on organ function related to pharmacology. However, consistent with the view of other OSWG subcommittees, the use of analogues should be approached with caution, particularly due to concerns about potential irrelevant off-target effects with a molecular entity that may have a different sequence, length, and/or chemistry than the clinical candidate.

Selection of evaluation time points and interdose intervals should to reflect the pharmacodynamics (PD) and pharmacokinetics (PK; including tissue persistence) of the ON as with any drug candidate. The ideal window of time for capturing safety pharmacology endpoints would encompass the period during which plasma concentration, target tissue concentration and/or pharmacological activity are maximum. For most ONs, particularly those administered parenterally, these conditions can be closely approximated within the standard 24-hour post-dose monitoring period. However, for some ONs, optimal evaluation may be achieved with an extended monitoring period or within the context of repeat-dose administration.

In selecting the species for ON safety pharmacology studies, the Subcommittee recommends that the following points be taken into consideration:

1. The relevance of the model. Specifically, the extent to which the model has been “validated” to be predictive of effects in humans, as well as the historical precedent for use in ON programs or more generally for safety pharmacology investigations. The model should be appropriately sensitive to the parameters being evaluated, and the response(s) should be reproducible. These criteria are no different from those for any other class of therapeutics.

2. The species used for other aspects of the nonclinical development program. Use of the same species for pharmacology, pharmacokinetic, and toxicity evaluations provides continuity and allows the possibility of a more integrated, mechanistic understanding of toxicity/PK/PD relationships. When NHPs are not being used for toxicity testing, use of the NHP for the safety pharmacology evaluation may be advisable, particularly for the cardiovascular evaluation if the ON is in a structural subclass that has been associated with complement activation.

3. The responsiveness of the species to the intended pharmacologic activity of the ON. Effects on organ function could occur due to the mechanism of action. Thus, use of a pharmacologically responsive species may be advantageous for detecting mechanism-based effects as well as those related to off-target binding or chemistry. If a nonresponsive species is selected, the sponsor should acknowledge that the study cannot evaluate mechanism-based effects.

4. The similarity in target distribution between the species and humans. Ideally, the species selected should mimic the target distribution in humans, if known, in order to provide the most clinically relevant results. However, this information is often not available, especially at an early stage in development.

In the Subcommittee's opinion, “best practice” is the selection of a pharmacologically relevant species with target distribution similar to that in humans. Regardless of the species chosen, the rationale for the selection, as well as the limitations, should be included in the regulatory submission.

Organ Systems

Cardiovascular

The assessment of cardiovascular safety for ON-based therapeutics is typically a key part of the overall safety pharmacology evaluation for such products. Several considerations regarding the evaluation of cardiovascular safety pharmacology that are specific for ONs are discussed below.

In vitro cardiovascular safety studies

In vitro studies to assess the risk of cardiovascular effects typically focus on characterizing the potential to cause electrocardiographic (ECG) disturbances. The hERG assay evaluates test article effects on the repolarizing potassium current that is most commonly implicated in drug-induced QT prolongation. The hERG assay is a primary component of an integrated risk assessment of the potential for delayed ventricular repolarization by human pharmaceuticals. This test is recommended by the ICH (ICH S7B, 2005) and has been adopted by worldwide regulatory bodies. The importance of the hERG assay for small molecules is based on a large body of published literature that establishes a clear, albeit not perfect, relationship between drug-related QT prolongation in humans and inhibition of the hERG channel current in vitro for small molecule drugs (Redfern et al., 2003; WALLIS, 2010; Pollard et al., 2010; GINTANT, 2011). In addition, interactions between various small molecules and structural determinants within the channel deep pore have been characterized in detail (Vanderberg et al., 2012; Perry et al., 2010; references therein). For larger molecules such as biotherapeutics, the lack of published reports for these types of interactions, as well as considerations regarding their size and inability to have intracellular access, have provided a widely accepted scientific rationale for why testing these molecules in the hERG channel assay lacks value (Vargas et al., 2008, 2013). However, ONs have distinct physicochemical and biological properties that distinguish them from both small molecules and biotherapeutics (Nakazawa et al., 2008).

For ONs, testing in the hERG assay has been performed mostly for the PS class of ONs. As a whole, data suggest that the hERG channel is unaffected by PS ONs because no significant inhibition is detected at concentrations ranging from 100 to 300 μM (free ON concentration), which is typically 1,000-fold over the anticipated maximum plasma concentration (C_max) in the clinic (Kim et al., 2013). Thus, PS ONs do not appear to interact directly with the hERG channel and, therefore, testing of this type of ON in this assay is not recommended.

Furthermore, testing may not be warranted, even for ONs that contain novel chemical modifications, when:

• Sufficient data are available from in vivo studies to document the absence of a potential for the ON to directly cause ECG disturbances (e.g., robust data from a study employing implanted or jacketed telemetry in non-rodents);

• Systemic exposure is limited, and concern about the likelihood of ECG abnormalities in relation to the pharmacologic activity of the ON is low;

• Intracellular access to the hERG channel in the test system cells is unlikely (e.g., for an aptamer conjugated to polyethylene glycol (PEG) or an ON with a receptor-specific conjugate); or

• The ON is part of a complex formulation and/or when the delivery system contains an excipient(s) that is likely to affect the validity of the test.

Because these conditions are generally true for many types of ONs, the testing of ONs for inhibition of hERG current is not warranted a priori, but rather, it is recommended only in specific circumstances. For example, an appropriate use of this test might be to follow up on observations of altered cardiovascular function in an in vivo study. Other circumstances may include a role of the ON target gene in cardiovascular function, known or predicted accumulation in heart tissue, novel chemistry or other structural attributes that might predispose towards cardiovascular effects, and similar concerns about the formulation or delivery system, if applicable. In general, in vivo assessments, as prescribed in the ICH S7A guideline and detailed in the sections that follow, are recommended as the initial and, in most cases, sufficient means for ON cardiovascular evaluation.

In vivo cardiovascular safety studies

The primary objective of in vivo safety pharmacology studies has traditionally been the assessment of acute effects after single-dose administration. For PS and other highly polyanionic ONs, the most commonly observed changes in cardiovascular parameters are hemodynamic effects stemming from complement activation (Galbraith et al., 1994; Henry et al., 1997a). Similar hemodynamic changes have not been reported for other classes of ONs; in addition, other changes (including those related to hERG inhibition) are rarely seen with any class of ONs.

NHPs, mainly cynomolgus and rhesus monkeys, have traditionally been used for conducting in vivo safety pharmacology studies with most subclasses of therapeutic ONs. This precedent was established based mainly on the potential for PS ONs (and other highly polyanionic ONs) to activate the alternative pathway of complement in NHPs, which translated into hemodynamic alterations, mainly hypotension (Henry et al., 1997a). This effect is related to blood (plasma) concentration and occurs mainly upon bolus intravenous injection or short-duration intravenous infusion, although activation can be triggered by other routes of administration (e.g., subcutaneous injection) at high dose levels (Henry et al., 1997a). These findings with first-generation PS ONs served as the impetus for using NHPs for nearly all ON safety assessment programs, regardless of the backbone chemistry. Although other chemical backbones (e.g., phosphorodiamidate morpholino) may not have demonstrated such effects, published information on them is sparse (e.g., Sazani et al., 2010), so they should be subjected to the same tests. The use of NHPs for cardiovascular safety testing for all ONs containing a charged backbone was recommended in several publications by U.S. Food and Drug Administration (FDA) pharmacology/toxicology reviewers (Ahn and DeGeorge, 1998; Black et al., 1994), and the practice was subsequently widely adopted. However, in vitro data (Levin et al., 2001), as well as a growing body of in vivo data (Isis Pharmaceuticals, unpublished data), suggest that the dose-response for complement activation in NHPs may overpredict human sensitivity. For those ONs that do not contain a modified polyanionic backbone structure or that would otherwise not be expected to induce complement activation, the use of other nonrodent species (e.g., dog or minipig) for cardiovascular safety assessment could be considered. In fact, new data suggest that the minipig may also be an acceptable species for assessment of complement-mediated cardiovascular effects (Szebeni et al., 2012; Dézsi et al., 2013); however, to the Subcommittee's knowledge, no data are available regarding ON-related effects in this model.

Two basic strategies have been employed for obtaining cardiovascular safety data with ONs in NHPs or other nonrodent species. Conducting a stand-alone study has been the traditional approach, but more recently, investigators have incorporated cardiovascular safety assessment into repeat-dose toxicity studies. According to the experience of the Subcommittee members, the most common study design for a stand-alone cardiovascular safety pharmacology study in NHPs entails the use of at least four animals with surgically implanted radiotelemetry transmitters and employs a rising-dose design with appropriate washout periods between doses. This escalating-dose design is often preferable to the Latin square design because these safety pharmacology studies are typically conducted early in the program (sometimes prior to or concurrent with the repeat-dose studies conducted to support filing of an Investigational New Drug (IND) application), and uncertainty often exists about whether the highest dose level may cause a potentially lethal degree of complement activation or other adverse effect. However, a Latin square design may be acceptable, particularly for non-PS ONs, if tolerability of the ON has been characterized at the doses selected for the safety pharmacology study. For instance, a single-dose or dose range-finding toxicity study or PK/PD study prior to the safety pharmacology study will inform about tolerability and may also provide information about the appropriate telemetry monitoring period. In general, for PS ONs, the dose-response in NHPs for complement activation and downstream cardiovascular sequelae are somewhat predictable based on prior experience with structurally related ONs, although some outliers with dose responses shifted high or low have been encountered.

The telemetry devices used most commonly can continuously transmit data on blood pressure, heart rate and ECG (lead 2) activity, as well as body temperature. As desired, blood sampling can be incorporated into the study for toxicokinetic (TK) and/or complement split product analyses to characterize the interrelationships among exposure (plasma concentration), acute activation of the alternative complement pathway, and hemodynamic alterations. In the stand-alone type of study, incorporation of other safety pharmacology endpoints, such as evaluation of neurologic and respiratory function, may also be a design option. However, caution should be used to not overload these studies with procedures that are stressful to the animals, as stress can compromise the main objective of collecting cardiovascular telemetry data from conscious, undisturbed animals. Therefore, collection of blood samples for TK, complement, and other parameters should be coordinated by careful selection of appropriate common time points; alternatively, samples can be collected in a dedicated round of dosing with the same animals (following an appropriate wash-out period) or from the toxicity study animals (if the studies are conducted at the same dose levels). The excitement-induced disruption of blood pressure and heart rate associated with blood sampling or other restraint procedures typically subsides within 15 to 60 minutes after human interaction, so sparse sampling can be incorporated.

An alternative approach is to incorporate the cardiovascular assessment into a repeat-dose toxicity study, which allows monitoring of effects following both single and repeated doses. This can be accomplished through the use of telemetry in a subset of study animals to enable monitoring of cardiovascular parameters in conscious, unrestrained animals. Although animals can be implanted with telemetry devices as described above, this is an invasive surgical procedure that must be properly managed in the context of a toxicity study. More recently, jacketed external telemetry (JET) systems have been validated; these systems involve the use of surface electrodes for ECG and minimally invasive insertion of a femoral artery catheter coupled with subcutaneous implant of the device for continuous unrestrained hemodynamic monitoring (Chui et al., 2009; Vargas et al., 2013). The advantages and disadvantages of implanted vs. jacketed telemetry are described by Vargas et al., 2013. The use of peripheral cuff blood pressure measurements is not recommended, as the measure requires restraint, lacks sensitivity to detect subtle changes in both pressor and depressor responses, only captures a snapshot of data in time, and is an indirect measure of blood pressure. Regardless of whether cardiovascular safety is assessed by conventional implanted telemetry or via JET, the recordings should not be scheduled on days with other stressful study activities, such as intensive TK sample collection or other procedures that can disturb the animals, which can introduce some constraints into the study design.

The Subcommittee does not have a strong recommendation about whether stand-alone studies or incorporation into repeat-dose toxicity studies is optimal; this judgment should be made on a case-by-case basis. Considering that the primary concern about cardiovascular safety with many ONs, most notably PS ONs, has been the potential for complement activation, which is an acute effect, a traditional stand-alone single-dose study may be regarded as adequate (Henry et al., 2002). However, if there is cause for concern about other possible cardiovascular disturbances, especially those that might manifest upon repeated dosing, the incorporation of cardiovascular safety endpoints into the nonrodent repeat-dose toxicity study may have greater value. It should be noted that this approach typically requires surgical preparation of 12 or more telemeterized animals (usually ≥two per sex per group in three dose groups, plus controls) as compared with a total of four animals that are typically used for a stand-alone study. Hence, although the data derived from this strategy may be more robust than can be obtained from a conventional single-dose safety pharmacology study, the resource allocation is more substantial. On the other hand, the leveraging of toxicity study animals to derive cardiovascular safety data obviates the need for an additional in vivo study, with the attendant use of additional animals.

As mentioned, the use of telemeterized animals in either a stand-alone study or incorporated into a repeat-dose toxicity study typically enables continuous monitoring of blood pressure, heart rate, and ECG activity over a designated post-dose time interval. The recording period is often 24 hours or longer (on each recording occasion), depending on the time frame for predicted peak deposition in tissues and/or maximal PD activity. Therefore, the Subcommittee agreed that the data derived from a well-designed in vivo cardiovascular safety study is sufficiently comprehensive and may obviate the need for in vitro testing for effects on hERG channel current.

Central nervous system

For the purpose of this discussion, the peripheral and central nervous system will be referred to simply as the CNS. Historically, concern regarding CNS safety of ON-based therapeutics has been minimal because systemically delivered ONs, particularly those with charged backbones, are cleared rapidly from the blood and typically do not cross the blood–brain barrier (BBB) to a quantitatively significant extent (Boado et al., 1998). In support of this low level of concern, systemically administered ONs that have advanced into clinical development have not been shown to produce CNS pathology or serious CNS adverse events (Kim et al., 2013). However, the absence of pathology does not mean that ONs are incapable of producing behavioral effects. In fact, nuclease-resistant ONs such as PS ONs and peptide nucleic acids can penetrate the BBB with sufficient exposure to affect CNS function (Banks et al., 2001; Tyler et al., 1999). Furthermore, it has been reported that PS ONs cross the BBB via a saturable transport system, the oligonucleotide transport system-1 (Banks et al., 2001); however, more work needs to be performed to substantiate this claim. Therefore, although the risk of adverse effects on the CNS is relatively low compared with other organs, the potential for ONs to induce CNS safety issues should not be ignored and must be evaluated according to ICH S7A guidelines. Additionally, the peripheral nervous system is exposed to ONs, and its function could also be affected.

The traditional approach to CNS safety pharmacology evaluations is a stand-alone neurofunctional assessment in rodents, which has been the gold standard in the evaluation of CNS safety pharmacology endpoints for new chemical entities; hence, it is an appropriate option for ONs. It consists of performing a functional observational battery (FOB) (Moser and Ross, 1996) or Irwin test (IRWIN, 1968) at one or more time points following single-dose administration of the test article. In addition, although not required, many sponsors perform a quantitative measure of locomotor activity as part of the neurofunctional assessment. This type of evaluation permits the uninhibited assessment of the acute effects produced by a test article on central and peripheral nervous system function. However, this approach involves the use of 18 to 32 animals, depending on the number of treatment groups (3 to 4) and treatment group size (6 to 8), and increases the amount of bulk test article required for the project compared to the approaches listed below. In addition, this approach typically assesses only the acute effects of test article administration and provides no information on possible effects that may occur after repeat dosing. Therefore, consideration should be given to assessing the effects of an ON after repeat dosing. In any case, in the event that a systemically delivered ON is expected to achieve high exposure to the CNS and/or is expected to cause acute CNS effects, the Subcommittee recommends that a stand-alone neurofunctional assessment be performed.

A second approach for CNS safety pharmacology assessments is to incorporate a neurologic evaluation into a comprehensive, single-dose safety pharmacology study in NHPs that would also include cardiovascular and respiratory assessments. Neurological examinations are commonly performed on NHPs and typically include a basic assessment of behavior, cranial nerve function, and a few simple reflexes (PORSOLT, 2013). This approach of combining safety pharmacology assessments (i.e., cardiovascular, CNS, and respiratory) into one study is beneficial in terms of reducing the number of animals and test articles needed to assess safety pharmacology endpoints. One disadvantage to this type of study is that a compromise has to be made in terms of the timing of the cardiovascular, CNS, and respiratory assessments, as they cannot be measured concomitantly without affecting the integrity of the endpoints. More specifically, performing a neurological evaluation at scientifically justified time points would compromise cardiovascular endpoints due the stress induced by the examination. On the other hand, performing a neurological evaluation at an arbitrary time point post dose in order to capture cardiovascular endpoints at scientifically appropriate times calls into question the value of the neurological data. Another disadvantage, similar to the stand-alone assessment, is that only the acute effects of the ON are assessed in the combined safety pharmacology study. It may be desirable to investigate the potential for CNS effects over time.

Thus, a third approach is to incorporate the CNS evaluation into the repeat-dose toxicity study. In NHPs, the CNS evaluation would be as described above for the combined safety pharmacology assessment (i.e., neurologic examination); whereas in rodents, the evaluation would comprise an FOB/Irwin test as is done in a stand-alone study. This approach allows the assessment of CNS function after acute administration of the ON. Additional CNS assessments may be performed at a late stage in the dosing period of the toxicity study to evaluate the effect of the ON on CNS parameters over time and when steady-state exposure has been achieved. In addition, evaluations during the recovery period may be made in the event that CNS signals were observed during the dosing phase. The advantages of this approach include the opportunity for CNS evaluations after both single and repeat ON administration; interpretation of any signals within the context of the toxicokinetic results, as well as any other toxicities or pathology observed; and reduction in the number of animals and test article bulk needed to prosecute the safety strategy.

The Subcommittee does not have a strong recommendation about which approach is optimal; judgment should be made on a case-by-case basis. With any of these approaches, use of a pharmacologically relevant species is preferable, which will often lead to the NHP as the most appropriate species.

Of note, both the neurological exams in NHPs and the FOB/Irwin test in rodents assess basic neurological function. Many CNS adverse events that are observed in patients, such as cognitive dysfunction, psychiatric disturbances, disruption of circadian rhythms, and sensory or motor disturbances (e.g., peripheral neuropathies, paresthesias, myalgias, etc.), will not be detected by a neurological exam or FOB/Irwin test, nor do they typically present themselves during first-in-human studies. With such events, a thorough risk assessment should be performed based on scholarship around the ON target, including the role of the target in various brain processes, mechanism of action, and CNS expression. In addition, the clinical indication, patient population, and PK of the test article should be taken into account. From this information, the potential CNS effects may be predicted, and the level of risk for the production of these effects by the ON therapeutic may be determined. If the predicted CNS effect is judged to be a liability or risk to the project, the Subcommittee recommends that a specialized CNS study be designed to address the safety issue. An example of this type of approach would be the use of 24-hour continuous monitoring via electroencephalography and activity to assess alterations in circadian rhythms produced by a test article.

Finally, an abuse potential assessment that includes nonclinical and clinical components will be needed for any ONs that target the CNS (FDA Guidance on Assessment of Abuse Potential, 2010). Such an assessment may also be needed for other ONs in the unlikely event that (1) the ON produces unintended effects on the CNS, (2) the ON is pharmacologically similar to other drugs with known abuse potential, or (3) the ON produces psychoactive effects such as euphoria, mood changes, or sedation. The types of assessments and information that would be required, as well as the timing of the assessments relative to clinical development, are no different for ONs than for small molecule drugs.

Respiratory

An evaluation of respiratory safety pharmacology is considered appropriate for inclusion in most IND applications for parenterally administered products, as per ICH S7A guidelines. In general, the level of concern for pulmonary effects related to parenterally administered noninhaled ONs is considered to be very low. Although complement activation after ON administration has been shown to affect cardiovascular parameters in NHPs (Henry et al., 1997b) and has the potential to alter respiration, the Subcommittee is unaware of any direct effects on pulmonary function with any of the various parenterally delivered ON chemistries in development. On the other hand, the level of concern over pulmonary effects for inhaled ONs is understandably greater, in part due to potential irritant and inflammatory effects observed with inhaled formulations (Forbes et al., 2011; OWEN, 2013). With repeated exposure to inhaled ONs, inflammation associated with particulate accumulation and degradation and/or pharmacological effects may also occur (Guimond et al., 2008; Templin et al., 2000; Ali et al., 2001). Thus, the respiratory safety pharmacology strategy for inhaled ONs may be different from that for parenterally administered ONs. The recommended strategies for these different routes of administration are discussed separately below.

Typically, the most relevant route of administration is the intended route for clinical administration, which for most ONs is the intravenous or subcutaneous route. For non-rodent species, four animals per group are usually considered a sufficient sample size for detection of respiratory effects in well-acclimated animals; a larger group size is preferable when concern has been raised regarding potential pulmonary effects. Group sizes of six to eight are preferred for rodent studies. In all cases, unanesthetized animals should be used for respiratory assessment due to respiratory depressant effects of most anesthetics.

For assessment of respiratory safety, ICH S7A suggests measurement of frequency of breathing and at least one additional respiratory parameter. In rodents, these measurements are typically made in head-out or whole-body plethysmographs, which allow measurement of frequency of breathing, tidal volume, and their product, minute volume. Similar methods can be applied to non-rodent species. At some laboratories, non-rodent assessment may be limited to physical examination parameters (respiration rate, in addition to heart rate, body temperature, and blood pressure) conducted by a veterinarian. Physical exam data may be supplemented by the non-invasive determination of oxygen saturation or by invasively collecting blood for blood gas determination. However, like physical exams, these two methods are particularly insensitive parameters for the assessment of respiratory effects in otherwise healthy animals due to the tremendous dynamic reserve of the lung and because of the high variability inherent in these measurements. Therefore, although not recommended, if physical exams, oxygen saturation and/or blood gas analysis are chosen for the evaluation of respiratory safety in the non-rodent species, selection of an experienced laboratory that can achieve low variability in these parameters is important.

When concern exists regarding the ON's potential to produce respiratory effects, the Subcommittee does not recommend relying on physical exams to measure respiratory rate or blood gas evaluation. As per ICH S7A, “Clinical observation of animals is generally not adequate to assess respiratory function, and thus these parameters should be quantified by using appropriate methodologies.” The Subcommittee recommends that tidal volume should be measured along with frequency of breathing, allowing the calculation of minute ventilation. These two parameters are best measured using a pressure or flow plethysmograph or a pneumotachometer in larger animals. Unfortunately, plethysmography, or the use of a pneumotachagraph, requires physical or chemical restraint of the test subject; however, these techniques are still the gold standard for respiratory measurement. Animals must be acclimated to the restraining equipment to obtain reliable measurements with low variability within and between subjects. A newer, less invasive technique for use in rodent species is whole-body plethysmography. Recently, the ability to measure respiration in freely moving non-rodent species using jacketed (respiratory inductive plethysmography) or surgical transducer implantation (impedance plethysmography) has become available. However, the high variability associated with movement artifacts limits the sensitivity and usefulness of these techniques.

The strategy for performing respiratory safety pharmacology can be approached in several ways, each having different relative advantages. The options include (1) a dedicated acute respiratory safety study, (2) an acute study combining several safety pharmacology endpoints, or (3) incorporation of the respiratory safety pharmacology assessment into repeat-dose toxicity studies. Although a stand-alone assessment is always an acceptable option, the Subcommittee believes that one of the latter two methods may be scientifically sound, in addition to reducing the use of animals and study cost. One strategy that has been used successfully is incorporation of respiratory parameters into a comprehensive (cardiovascular, respiratory and neurobehavioral) stand-alone safety pharmacology study in telemetered NHPs. In this case, use of a pharmacologically responsive species is of added benefit. The addition of plethysmography to a cardiovascular safety pharmacology study in NHPs or dogs is feasible and does not add considerable expense; thus, such a combined study reduces the number of animals needed and allows the correlation of respiratory effects with cardiovascular effects. However, because this procedure is usually conducted in restrained animals, the restraint can interrupt the evaluation of cardiovascular parameters in undisturbed animals. This disruption can be minimized by conducting the pulmonary function testing for brief periods (15–30 minutes of acclimation to the restraint followed by at least 5 minutes of data recording) and limiting the frequency of this evaluation over each 24-hour monitoring period. If concern exists regarding the pharmacological target (especially if cardiovascular) or route of administration (e.g., inhalation), a combination safety assessment may not be appropriate.

The other approach, when concern about the ON target mechanism, route of administration and/or formulation is low, is to incorporate respiratory safety pharmacology parameters into repeat-dose toxicity studies. Typically, measurements are made on three occasions, prior to initiation of treatment, after the first—or first few—dose(s), and close to the end of the study. Prior acclimation of the animals to the required restraint procedure is necessary. A single measurement period of at least 5 minutes after the animal has acclimated to the restraint procedure is recommended. Best practice is for this measurement period to coincide with the approximate time of maximal ON concentration in the blood or target organ. Although this approach is acceptable for oncology products and biotherapeutics as of the publication of ICH S6(R1) and ICH S9, it may not always be acceptable to regulatory authorities for new chemical entities (including ONs) falling under ICH M3(R2) for indications outside of oncology. Members of the Subcommittee reported case examples of regulatory acceptance of this strategy, an approach that appears to be gaining broader acceptance (Authier et al., 2013) for the development of ONs.

Respiratory assessment in the context of a repeat-dose study may also be the best approach to assess pharmacology-related effects that may target the respiratory system and is recommended if such effects are of concern. This approach is also valuable when concern exists regarding cumulative nonspecific effects in the respiratory tract as may occur with inhalation of ONs. As above, the measurement of respiratory rate and tidal volume are preferred and may be feasible in such studies (using plethymosography in rodents and a pneumotachometer in non-rodents). Assessment in both stand-alone and repeat-dose toxicity studies may be warranted for compounds of concern based on pharmacology, route of administration, or an observed safety signal (e.g., lung histopathology findings) from a prior study. However, this approach is in excess of the guidance set forth in ICH S7A.

Inhaled ONs are under development and at least one has made it to phase 2 clinical studies. However, inhaled delivery of ONs poses special challenges, most prominently, the determination of inhaled dose. Such assessments require specialized procedures for determining airborne concentration and particle size, which must be in the respirable range (1 to 3 microns). Achieving sufficiently high concentrations or doses to assess large multiples of the clinical dose is typically not a problem for parenteral routes of administration but can be a challenge for inhaled routes. Animal welfare issues regarding the safe duration for which animals can be restrained physically during inhalation exposure may limit achieving sufficiently high exposures for safety assessment. In certain cases where duration is an issue, the inhalation dose may be supplemented with parenteral dosing to achieve adequately high systemic exposures. Furthermore, it may be arguable that a parenteral route (typically intravenous) will provide a better assessment of the adequacy of cardiovascular or nervous system function than would occur with inhalation exposure. As for species selection, in many cases ONs are tested in NHPs and rodents (mice or rats). These species are acceptable for inhalation studies; however, due to anatomical, logistical, and physiologic reasons, as well as the wealth of historical data held by regulatory agencies, when feasible, rats and dogs are the preferred species for inhalation studies. Further discussion about the specific pulmonary safety issues of inhaled ONs and approaches to their assessment is covered in a recent publication by the OSWG Inhalation Subcommittee (Alton et al., 2012).

Supplemental safety pharmacology studies

Renal/urinary system

The Subcommittee recommends that the evaluation of the renal and urinary systems be conducted as part of a repeat-dose toxicity study; this recommendation is consistent with ICH S7A guidance.

The level of concern for adverse effects on renal function is related primarily to an ON's propensity to distribute to and accumulate in the kidney. In general, ON therapeutics that are unconjugated, non-PEGylated, and delivered as simple formulations (e.g., saline) are filtered through the glomerulus and reabsorbed by the proximal tubular epithelium. In the case of PS ONs, localization and subsequent accumulation of the drug in the proximal tubules of the kidney occurs following repeat treatment. PS ONs may accumulate in proximal tubule cells in a dose-dependent manner, and tissue concentrations plateau as steady-state exposure is achieved. Therefore, the kidney may be a target organ of toxicity of systemically administered ON therapeutics. However, the extent of tissue distribution and accumulation in the kidney varies substantially within and across chemical classes.

Renal safety pharmacology evaluations are traditionally conducted after single-dose administration. For ON therapeutics, renal safety pharmacology is more appropriately assessed in a repeat-dose study because the potential effects on the kidney result primarily from the presence and persistence of drug in this tissue. Renal safety assessment can be conducted in rodents and/or non-rodents as part of the general toxicity evaluation; pharmacological responsiveness to the test article is not required unless the pharmacological target is in the kidney. Of note, the rat appears more sensitive to PS ON-related effects in the kidney relative to other species, including the mouse (Isis Pharmaceuticals, unpublished data). This sensitivity is hypothesized to be related to the susceptibility of rats to age-related nephropathy, a common spontaneous background lesion in this species (Hard et al., 2013). Because evaluation of renal safety pharmacology for ON therapeutics in rats may be confounded by this finding, especially in chronic studies, caution is advised when selecting the rat for evaluating ON-related renal effects.

After repeated administration of relatively low dose levels of ONs to rodents or non-rodents, changes in the kidney are typically limited to the presence of basophilic granules and/or cytoplasmic vacuoles in the proximal tubules. These findings are indicative of the presence of the ON and/or closely related metabolites within phagolysosomes of cells. In general, these findings are not associated with elevations in serum chemistry markers of kidney damage (e.g., blood urea nitrogen, creatinine, γ-glutamyl transpeptidase) or microscopic changes to the parenchyma and, therefore, are not considered adverse (Henry et al., 2000). Data on the more recent biomarkers (e.g., cystatin C, kidney injury molecule-1) are limited, and their predictive value has not been established for ONs.

Toxicologically relevant and potentially adverse changes may be observed in the kidney after repeated administration of relatively high ON dose levels. For PS ONs, these findings may include proximal tubular epithelium apoptosis and degeneration/regeneration (Monteith et al., 1999; Henry et al., 2012). Affected tubules may have enhanced cytoplasmic basophilia, single cell necrosis, and/or loss of individual tubular cells (Frazier et al., 2013). These findings are considered to be secondary to ON accumulation. PS ON-related tubulotoxicity appears to be more prevalent in NHPs than mice because greater kidney concentrations can be achieved in NHPs (Monteith et al., 1999); however, this species difference is not necessarily evident for all ONs.

As a whole, microscopic changes in the kidney, including tubular epithelial cell degenerative findings, are often not associated with a deficit in kidney function. For example, after administration of a PS ON that produced mild tubular degenerative changes in NHPs, no changes were evident in glomerular filtration rate, glucose reabsorption, or urinary concentrating ability following water deprivation (Henry et al., 2012). In some cases, an increase in urinary protein/creatinine (P/C) ratio was observed in animals with minimal degenerative changes in the proximal tubules. Proteinuria is related to decreased tubular absorptive capacity (Frazier and Seely, 2012) resulting from tubular ON accumulation. Thus, P/C ratio and/or urinary albumin may be included as a parameter for safety assessment for this class of molecules in non-rodent safety studies. Any effects should be monitored for reversibility, which generally occurs in a manner consistent with gradual ON clearance from the kidney (Henry et al., 2008). A no-dose recovery period of several weeks is often necessary to detect a decrease in effect; however, the exact duration required will depend upon the half-life of the ON in the kidney.

In summary, evaluation of effects on the renal and urinary systems for any class of ON therapeutics is best characterized in repeat-dose general toxicity studies rather than in single-dose safety pharmacology studies. This recommendation is based primarily on findings associated with accumulation of ON in the kidney following repeat-dose treatment. Parameters to be examined include serum chemistry (urea, creatinine, albumin, total protein) in rodent and non-rodent studies and urinalysis (urinary volume, specific gravity, osmolality, pH, color, clarity) and urine chemistry [proteins (especially urinary albumin), creatinine, and protein/creatinine ratio] in non-rodent studies. These parameters, along with microscopic evaluation of the kidney in both species, should be adequate to assess the safety of ONs in the kidney and urinary system (EMEIGH HART, 2005). However, sponsors are encouraged to consider the assessment of kidney injury biomarkers for exploratory purposes as this will build the database regarding their predictive value for ONs.

Gastrointestinal system

The gastrointestinal (GI) system is not considered a core organ system and, thus, is considered of less concern per ICH S7A guidelines. This low level of concern is particularly true for the ONs, which are typically not administered orally. Furthermore, systemically administered ONs are rarely associated with adverse GI effects.

In the Subcommittee's view, a stand-alone safety pharmacology study to assess GI function is generally not warranted for ONs. Parameters routinely included in rodent or non-rodent single- or repeat-dose toxicity studies are sufficient to provide information on potential effects on GI function. Findings indicative of such an effect include body weight loss, decreased food consumption, clinical observations of emesis or qualitative changes in stool, and serum chemistry changes in alkaline phosphatase, phosphorus, and potassium. Although individual signs and symptoms are not sufficiently specific, a constellation of these findings generally provides enough information on any changes in the function of the GI system.

Specific GI safety pharmacology studies could be conducted in cases of heightened concern. Such concern could be triggered by GI signs and symptoms in the repeat-dose toxicity studies, on-target GI pharmacology, or a novel oral delivery technology. Additionally, when the clinical indication is a GI disease or the patient population is at risk for specific GI effects, safety pharmacology or investigative studies may be considered. In such cases, a range of both in vitro and in vivo methods/assays is available for selection depending on the GI liabilities (Harrison et al., 2004). Investigative methods related to function include quantitative stool count (such as included in a FOB), GI motility assessment (using charcoal, beads, or wireless capsule), absorption assessment (primary GI or cultured cells and ex vivo GI tissue); in vivo imaging technologies such as endoscopy and wireless cameras. These assays provide knowledge and information of GI functions and their regulation. Generally speaking, selection of the assays and options for study designs would be similar to those for GI safety pharmacology studies used to assess small molecules.

Immune system

Pro-inflammatory effects in animals have been well described for ON therapeutics (mainly those with a charged backbone and persistence in tissues), so evaluation of the immune system may be warranted for these molecules. For ONs that are intended to be immunostimulatory by binding to specific Toll-like receptors, such effects will be evaluated during pharmacology studies. However, for other subclasses of ONs, immunomodulatory effects are unintended, and the sequence of each molecule will dictate the severity and character of the effects. Some sponsors choose to assess cytokine secretion either in vitro using human peripheral blood lymphocytes or in vivo during the candidate screening and selection process. Because the significance of any observed changes is difficult to interpret, evaluation of a “pro-inflammatory” response may be best conducted during repeat-dose toxicity studies via clinical observations, hematology evaluations, and tissue histopathology. A detailed discussion of the concerns and recommended testing approaches is being developed by the OSWG Immunomodulatory Subcommittee.

Notwithstanding the above recommendations, sponsors should be aware of potential pro-inflammatory effects when interpreting the results of safety pharmacology studies.

Receptor binding

Many ONs exhibit protein binding as a result of their anionic nature. For example, functional changes related to the binding of PS ONs to Factor H (Henry et al., 1997b) and the tenase complex (Sheehan and Lan, 1998) have been well documented. However, these effects, if they occur, can be monitored in vivo through measurement of complement split products and activated partial thromboplastin time. Broad panel screening for binding to receptors, channels, and kinases, etc. has been performed with a limited number of ONs. No generalized patterns of binding have emerged, and the Subcommittee is unaware of any situations in which screening revealed an important property of an ON. Furthermore, to date, receptor binding screens have not been requested routinely by the regulatory agencies for ON programs.

Nonetheless, such studies may have value in certain circumstances. For instance, for ONs whose mechanism of action is through protein binding (e.g., aptamers and CpG ONs), a directed screen should be conducted to document specificity as part of the pharmacological characterization. The Subcommittee recommends that such investigations be conducted prior to the first dose in humans and evaluate interactions not only with proteins (e.g., receptors) of similar activity but also with proteins of similar structure as the intended target. In addition, in the event that an ON exhibits a novel and unexplained toxicity, a broad screen may have value as part of the investigation. If cell-based assays are utilized, consideration should be given to the likelihood of cellular uptake for access to intracellular receptor binding sites. In addition, in reporting of safety margins relative to in vivo plasma concentrations for highly protein-bound ONs, care should be taken to compare free-to-free and/or total-to-total concentrations. If a signal is observed in binding experiments, a follow-up functional assay should be conducted to determine whether a functional consequence will result.

Other

As indicated in ICH S7A, effects of ONs on organ systems not otherwise evaluated should be assessed when there is reason for concern. The presence of the therapeutic target in a specific organ system may increase the organ's sensitivity to undesirable effects, which would engender a higher level of concern. However, for the standard organ systems (cardiovascular, CNS, respiratory), the purported ‘absence’ of the molecular target is not sufficient reason for omitting the organ system from the safety pharmacology evaluations. Tissue distribution data may also help inform about the level of concern regarding safety pharmacology effects in a specific organ system. For instance, the liver may be a potential target for toxicity due to tissue distribution of the ON; however, no specific studies are recommended, as adverse effects on the liver are generally assessed as part of the general toxicity studies. For ONs that demonstrate hepatotoxicity, additional investigative studies could be conducted to assess effect on hepatic function.

In addition, for ONs delivered directly to an organ system (e.g., via inhalation, intrathecal, or oral administration), more extensive assessment of effects on that organ system should be undertaken. Conversely, traditional safety pharmacology studies are not warranted for ONs that are delivered locally (e.g., intravitreally or topically) and for which systemic exposure is demonstrated to be negligible.

Conclusions

Oligonucleotides are highly specific agents that tend to demonstrate limited clinically relevant safety pharmacology concerns. However, one concern historically associated with systemically administered PS ONs and other ONs with highly polyanionic backbone structures was hemodynamic changes observed in NHPs triggered by alternative pathway complement activation. Because of this concern, the Safety Pharmacology Subcommittee of the OSWG has concluded that the primary emphasis for safety pharmacology studies of systemically administered ONs should be the evaluation of in vivo cardiovascular function, ideally in a pharmacologically responsive non-rodent species, preferably NHP. However, due to a species difference in sensitivity to complement activation, hemodynamic effects observed in the NHP may not translate to the clinic. The Subcommittee finds that, in general, in vitro hERG testing does not provide any additional value and is not warranted.

Limited to no effects on pulmonary or central nervous tissues have been noted for most types of ONs delivered systemically. Safety pharmacology assessment of CNS and respiratory function is recommended using the standard levels of rigor in the context of a stand-alone, combined function, or repeat-dose toxicity study to evaluate any major liabilities of the ON, unless guided differently by pharmacological mechanism, site of administration, or prior experience with the chemical class. Other dedicated safety pharmacology studies, beyond the core battery, generally have limited value for ONs. Evaluation of functional changes in the kidney and GI tract, as well as unintended “pro-inflammatory” effects, is best incorporated into repeat-dose toxicity studies. Additional studies, such as receptor- or ligand-binding profiling, should be considered only for investigative purposes.

In general, the Subcommittee's recommendations are consistent with those of the Safety Pharmacology Society (Leishman et al., 2012), as well as those brought forth by others (Vargas et al., 2008, 2013; Nakazawa et al., 2008; Hondeghem and De Clerk, 2012). The reader is advised that these recommendations are the consensus opinion of the Subcommittee and do not necessarily reflect the current expectations of regulatory authorities.

Footnotes

Acknowledgments

The authors would like to thank Hugo Vargas and the board of the Safety Pharmacology Society, Sandra Love, and Stephen Shrewsbury for their critical review of the manuscript.

Author Disclosure Statement

The submitted manuscript is a committee position document that does not promote any particular product(s), and no competing financial interests exist for any of the authors.

References

AHN

C.H.

, and DEGEORGE

J.J.

(1998). Preclinical development of antisense oligonucleotide therapeutics for cancer: regulatory aspects. In: Clinical Trials of Genetic Therapy with Antisense DNA and DNA Vectors. Wickstrom

, ed. (Marcel Dekker, Inc., New York), pp. 39–52.

ALI

, LEONARD

S.A.

, KUKOLY

C.A.

, JAMES METZGER

, WOOLES

W.R.

, MCGINTY

J.F.

, TANAKA

, SANDRASAGRA

, and NYCE

J.W.

(2001). Absorption, distribution, metabolism, and excretion of a respirable antisense oligonucleotide for asthma. Am. J. Respir. Crit. Care Med., 163, 989–993.

ALTON

E.W.

, BOUSHEY

H.A.

, GARN

, GREEN

F.H.

, HODGES

, MARTIN

R.J.

, MURDOCH

R.D.

, RENZ

, SHREWSBURY

S.B.

, SEGUIN

, et al. (2012). Clinical expert panel on monitoring potential lung toxicity of inhaled oligonucleotides: consensus points and recommendations. Nucleic Acid Ther., 22, 246–254.

AUTHIER

, VARGAS

H.M.

, CURTIS

M.J.

, HOLBROOK

, and PUGSLEY

M.K.

(2013). Safety pharmacology investigations in toxicology studies: an industry survey. J. Pharmacol Toxicol Methods, 68, 44–51.

BANKS

W.A.

, FARR

S.A

, BUTT

, KUMAR

V.B.

, FRANKO

M.W.

, and MORLEY

J.E.

(2001). Delivery across the blood-brain barrier of antisense directed against amyloid beta: reversal of learning and memory deficits in mice overexpressing amyloid precursor protein. J. Pharmacol. Exp. Ther., 297, 1113–1121.

BLACK

L.E.

, FARRELLY

J.G.

, CAVAGNARO

J.A.

, AHN

C-H.

, DEGEORGE

J.J.

, and TAYLOR

A.S.

(1994). Regulatory considerations for oligonucleotide drugs: updated recommendations for pharmacology and toxicology studies. Antisense Res. Dev., 4, 299–301.

BOADO

R.J.

, TSUKAMOTO

, and PARDRIDGE

W.M.

(1998). Drug delivery of antisense molecules to the brain for treatment of Alzheimer's disease and cerebral AIDS. J. Pharm. Sci., 87, 1308–1315.

CHUI

R.W.

, FOSDICK

, CONNER

, JIANG

, BRUENNER

B.A.

, and VARGAS

(2009). Assessment of two external telemetry systems (PhysioJacket™ and JET™) in beagle dogs with telemetry implants. J. Pharmacol. Toxicol. Methods, 60, 58–68.

DÉZSI

, SZÉNÁSI

, URBANICS

, ROSIVALL

, and SZEBENI

(2013). Cardiopaulmoary and hemodynamic changes in complement activation-related pseudoallergy. Health, 5, 1032–1038.

10.

EMEIGH HART

S.G.

(2005). Assessment of renal injury in vivo. J. Pharmacol. Toxicol. Methods, 52, 30–45.

11.

Food and Drug Administration (FDA). (2010.) Draft Guidance for Industry: Assessment of Abuse Potential of Drugs. (U.S. Department of Health and Human Services, Silver Spring, MD).

12.

FORBES

, ASGHARIAN

, DAILEY

L.A.

, FERGUSON

, GERDE

, GUMBLETON

, GUSTAVSSON

, HARDY

, HASSALL

, JONES

, et al. (2011). Challenges in inhaled product development and opportunities for open innovation. Adv Drug Deliv. Rev., 63, 69–87.

13.

FRAZIER

K.S.

, and SEELY

(2012). Urinary system. In: Toxicologic Pathology: Preclinical Safety Assessment. Sahota

P.S.

, Popp

J.A.

, Hardisty

J.F.

, and Gopinath

, eds. (CRC Press, Taylor and Francis Group, Boca Raton, FL), pp. 421–484.

14.

FRAZIER

K.S.

, SOBRY

C.E.

, DERR

, ADAMS

M.J.

, DEN BESTEN

, DE KIMPE

, FRANCIS

, GALES

T.L.

, HAWORTH

, MAGUIRE

S.R.

, et al. FRAZIER(2013). Species-specific inflammatory responses as a primary component for the development of glomerular lesions in mice and monkeys following chronic administration of a second-generation antisense oligonucleotide. Toxicol. Pathol. In Press. doi:10.1177/0192623313505781.

15.

GALBRAITH

W.M.

, HOBSON

W.C.

, GICLAS

P.C.

, and SCHECHTER

P.J.

(1994). Complement activation and hemodynamic changes following intravenous administration of phosphorothioate oligonucleotides in the monkey. Antisense Res. Dev., 4, 201–206.

16.

GINTANT

(2011). An evaluation of hERG current assay performance: translating preclinical safety studies to clinical QT prolongation. Pharmacol. Ther., 129, 109–119.

17.

GUIMOND

, VIAU

, AUBE

, RENZI

P.M.

, PAQUET

, and FERRARI

(2008). Advantageous toxicity profile of inhaled antisense oligonucleotides following chronic dosing in non-human primates. Pulm. Pharmacol. Ther., 21, 845–854.

18.

HARD

G.C.

, BANTON

M.I.

, BRETZLAFF

R.S.

, DEKANT

, FOWLES

J.R.

, MALLETT

A.K.

, MCGREGOR

D.B.

, ROBERTS

K.M.

, SIELKEN

R.L.

Jr. , VALDEZ-FLORES

, and COHEN

S.M.

(2013). Consideration of rat chronic progressive nephropathy in regulatory evaluations for carcinogenicity. Toxicol. Sci., 132, 268–275.

19.

HARRISON

A.P.

, ERLWANGER

K.H.

, ELBRØND

V.S.

, ANDERSEN

N.K.

, and UNMACK

M.A.

(2004). Gastrointestinal-tract models and techniques for use in safety pharmacology. J. Pharmacol. Toxicol. Methods, 49, 187–199.

20.

HENRY

S.P.

, MONTEITH

, KORNBRUST

D.J.

, and LEVIN

A.A.

(1997a). Effects of intravenous infusion of phosphorothioate oligonucleotides on coagulation, complement activation, and hemodynamics. Nucleosides Nucleotides, 16, 1673–1676.

21.

HENRY

S.P.

, GICLAS

P.C.

, LEEDS

, PANGBURN

, AULETTA

, LEVIN

A.A.

, and KORNBRUST

D.J.

(1997b). Activation of the alternative pathway of complement by a phosphorothioate oligonucleotide: potential mechanism of action. J. Pharmacol. Exp. Ther., 281, 810–816.

22.

HENRY

, STECKER

, BROOKS

, MONTEITH

, CONKLIN

, and BENNETT

C.F.

(2000). Chemically modified oligonucleotides exhibit decreased immune stimulation in mice. J. Pharmacol. Exp. Ther., 292, 468–479.

23.

HENRY

S.P.

, BEATTIE

, YEH

, CHAPPEL

, GICLAS

, MORTARI

, JAGELS

M.A.

, KORNBRUST

D.J.

, and LEVIN

A.A.

(2002). Complement activation is responsible for acute toxicities in rhesus monkeys treated with a phosphorothioate oligodeoxynucleotide. Int. Immunopharm., 2, 1657–1666.

24.

HENRY

S.P.

, KIM

T.W.

, KRAMER-STRICKLAND

, ZANARDI

T.A.

, FEY

R.A.

, and LEVIN

A.A.

(2008). Toxicologic properties of 2-O-methoxyethyl chimeric antisense inhibitors in animals and man. In: Antisense Drug Technology: Principles, Strategies and Applications. Crooke

S.T.

, ed. (CRC Press, Taylor and Francis Group, Boca Raton, FL), pp. 327–364.

25.

HENRY

S.P.

, JOHNSON

, ZANARDI

T.A.

, FEY

, AUYEUNG

, LAPPIN

P.B.

, and LEVIN

A.A.

(2012). Renal uptake and tolerability of a 2'-O-methoxyethyl modified antisense oligonucleotide (ISIS 113715) in monkey. Toxicology, 301, 13–20.

26.

HONDEGHEM

, and DE CLERCK

(2012). Preclinical cardiovascular safety evaluations of biologics: optimizing the approach and follow-up. BioDrugs, 26, 275–82.

27.

ICH (International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use) (2001). ICH Topic S7A: Guidance for Industry: Safety Pharmacology Studies for Human Pharmaceuticals. www.fda.gov/downloads/RegulatoryInformation/Guidances/ucm129156.pdf

28.

ICH (2005). ICH Topic S7B: Guidance for Industry: Nonclinical Evaluation of the Potential for Delayed Ventricular Repolarization (QT Interval Prolongation) by Human Pharmaceuticals. www.fda.gov/downloads/RegulatoryInformation/Guidances/ucm129122.pdf

29.

ICH. (2010a). ICH Topic M3(R2): Nonclinical Safety Studies for the Conduct of Human Clinical Trials and Marketing Authorization for Pharmaceuticals. www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm073246.pdf

30.

ICH. (2010b). ICH Topic S9: Nonclinical Evaluation for Anticancer Pharmaceuticals. www.fda/gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm085389.pdf

31.

ICH (2011). ICH Topic S6(R1): Preclinical Safety Evaluation of Biotechnology-Derived Pharmaceuticals. www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Safety/S6_R1/Step4/S6_R1_Guideline.pdf

32.

IRWIN

(1968). Comprehensive observational assessment: Ia. A systematic, quantitative procedure for assessing the behavioral and physiologic state of the mouse. Psychopharmacologia (Berl), 13, 222–257.

33.

KIM

T-W.

, KIM

K-S.

, SEO

J-W.

, PARK

S-Y.

, and HENRY

S.P.

(2014). Antisense Oligonucleotides on Neurobehavior, Respiratory, and Cardiovascular Function, and hERG Channel Current Studies. J. Pharmacol. Toxicol. Methods, 69, 49–60.

34.

LEISHMAN

D.J.

, BECK

T.W.

, DYBDAL

, GALLACHER

D.J.

, GUTH

B.D.

, HOLBROOK

, ROCHE

, and WALLIS

R.M.

(2012). Best practice in the conduct of key nonclinical cardiovascular assessments in drug development: current recommendations from the Safety Pharmacology Society. J. Pharmacol. Toxicol. Methods, 65, 93–101.

35.

LEVIN

A.A.

, TEMPLIN

M.V.

, HENRY

S.P.

, and MONTEITH

(2001). Toxicology of antisense oligonucleotides. In: Antisense Drug Technology: Principles, Strategies and Applications. Crooke

S.T.

, ed. (Marcel Dekker, New York), pp. 201–244.

36.

MONTEITH

D.K

, HORNER

M.J.

, GILLETT

N.A.

, BUTLER

, GEARY

, BURCKIN

, USHIRO-WATANABE

, and LEVIN

A.A.

(1999). Evaluation of the renal effects of an antisense phosphorothioate oligodeoxynucleotide in monkeys. Toxicol. Pathol., 27, 307–317.

37.

MOSER

V.C.

, and ROSS

J.F.

(eds.) (1996). Training Video and Reference Manual for a Functional Observational Battery. (American Industrial Health Council. Washington, DC)

38.

NAKAZAWA

, KUROKAWA

, KIMURA

, WAKATA

, HISADA

, INOUE

, SAGAMI

, HEIDEL

S.M.

, KAWAKAMI

, SHINODA

, et al. (2008). Safety assessment of biopharmaceuticals: Japanese perspective of ICH S6 guideline maintenance. J. Toxicol. Sci., 33, 277–282.

39.

OWEN

(2013). Regulatory toxicology considerations for the development of inhaled pharmaceuticals. Drug Chem. Toxicol., 36, 109–118.

40.

PERRY

, SANGUINETTI

, and MITCHESON

(2010). Symposium review: revealing the structural basis of action of hERG potassium channel activators and blockers. J. Physiol., 588, 3157–3167.

41.

POLLARD

, ABI GERGES

, and BRIDGLAND-TAYLOR

M.H.

(2010). An introduction to QT interval prolongation and non-clinical approaches to assessing and reducing risk. Br. J. Pharmacol., 159, 12–21.

42.

PORSOLT

R.D.

(2013). The usefulness of non-human primates in central nervous system safety pharmacology. J. Pharmacol. Toxicol. Methods, 68, 23–29.

43.

REDFERN

W.S.

, CARLSSON

, DAVIS

A.S.

, LYNCH

W.G.

, MACKENZIE

, PALETHORPE

, SIEGL

P.K.S.

, STRANG

, SULLIVAN

A.T.

, WALLIS

, et al. (2003). Relationships between preclinical cardiac electrophysiology, clinical QT interval prolongation and torsade de pointes for a broad range of drugs: evidence for a provisional safety margin in drug development. Cardiovasc. Res., 58, 32–45.

44.

SAZANI

, WELLER

D.L.

, AND SHREWSBURY

S.B.

(2010). Safety pharmacology and genotoxicity evaluation of AVI-4658. Int. J. Toxicol., 29, 143–156.

45.

SZEBENI

, BEDÕCS

, CSUKÁS

, ROSIVALL

, BÜNGER

, and URBANICS

(2012). A porcine model of complement-mediated infusion reactions to drug carrier nanosystem and other medicines. Adv. Drug Deliv. Rev., 64, 1706–1716.

46.

SHEEHAN

J.P.

, and LAN

H.C.

(1998). Phosphorothioate oligonucleotides inhibit the intrinsic tenase complex. Blood, 92, 1617–1625.

47.

TEMPLIN

, LEVIN

, GRAHAM

, ABERG

, AXELSSON

, BUTLER

, GEARY

, and BENNETT

(2000). Pharmacokinetic and toxicity profile of a phosphorothioate oligonucleotide following inhalation delivery to lung in mice. Antisense Nucleic Acid Drug Dev., 10, 359–368.

48.

TYLER

B.M.

, JANSEN

, MCCORMICK

D.J.

, DOUGLAS

C.L.

, BOULES

, STEWART

J.A.

, ZHAO

, LACY

, CUSACK

, FAUQ

, and RICHELSON

(1999). Peptide nucleic acids targeted to the neurotensin receptor and administered i.p. cross the blood-brain barrier and specifically reduce gene expression. Proc. Natl. Acad. Sci. U. S. A., 96, 7053–7058.

49.

VANDERBERG

J.I.

, PERRY

M.D.

, PERRIN

M.J.

, MANN

S.A.

, KE

, and HILL

A.P.

(2012). hERG K⁺channels: structure, function, and clinical significance. Physiol. Rev., 92, 1393–1478.

50.

VARGAS

, BASS

A.S.

, BREIDENBACH

, FELDMAN

H.S.

, GINTANT

G.A.

, HARMER

A.R.

, HEATH

, HOFFMANN

, LAGRUTTA

, LEISHMAN

, et al. (2008). Scientific review and recommendations on preclinical cardiovascular safety evaluation of biologics. J. Pharmacol. Toxicol. Meth., 58, 72–76.

51.

VARGAS

H.M.

, AMOUZADEH

H.R.

, and ENGWALL

M.J.

(2013). Nonclinical strategy considerations for safety pharmacology: evaluation of biopharmaceuticals. Expert Opin. Drug Safety, 12, 91–102.

52.

WALLIS

R.M.

(2010). Integrated risk assessment and predictive value to humans of non-clinical repolarization assays. Br. J. Pharmacol., 159, 115–121.