Abstract
Non-human primates (NHP) are used to best understand and address pharmacology and toxicology obligations for human patients with highest and/or unmet need. In order to ensure the most appropriate care and use of NHP, it is important to understand the normal micro flora and fauna of NHP and ensure their utmost health to generate the most valuable and applicable data. There are many infections, including viral, bacterial, parasitic, and fungal that may perturb physiologic endpoints relevant to human health, and are essential to monitor and/or eradicate for NHP health. This publication captures a discussion involving the experience, knowledge and opinion from academic, industry and government experts regarding emerging and normal infections in NHP as they relate to immunotoxicity, and treatment and consequences of known infections.
Introduction
Drug development requires a number of studies in rodent and non-rodent animals. Toxicological studies of small molecule therapeutics have conventionally used rodents and canines. However, with the recent increase in the number of highly specific human biologic therapeutics, non-human primates (NHP) are often the most appropriate non-rodent species for safety assessment studies. The use of NHP as test species has been common practice for both chemical and pharmaceutical companies for > 50 years (Vickers, Citation1969). It is important to be aware of the various types of clinical and latent NHP infections and their potential effects on toxicologically relevant endpoints.
The Immunotoxicology Technical Committee of the International Life Sciences Institute-Health and Environ-mental Institute (ILSI-HESI) held a workshop to discuss naturally occurring infections in NHP and immunotoxicity implications on October 22, 2008. This discussion was held to increase awareness of common and emerging NHP infections. There were three discussion sessions to review attendees’ experiences, strategies, and opinions regarding NHP infections including: (i) a general list or definition of infections of concern; (ii) treatment for infections; and (iii) whether infection status can be useful in assessing the degree of immunomodulation in study animals.
Discussion Session 1: Feasibility of ‘cleaner’ non-human primates for toxicology studies
Kenneth J. Olivier Jr., Joe Simmons, Keith Mansfield, and Nicholas Lerche
The intent of this session was to discuss whether non-human primates (NHP) used on toxicology studies in support of safety assessment for clinical trials in humans should be cleaner, meaning devoid of select pathogens not currently defined under specific pathogen-free (SPF). The overall conclusion was to use animals that would be most representative of the intended human population, which could mean many things as the discussion unfolded, but the general consensus was that it was not advisable to use ‘cleaner’ NHP, provided that animal welfare is not compromised. However, there is recent evidence at national primate research centers that newer breeding practices and research initiatives are trending toward reducing background infections and including additional pathogens in the definition of SPF animals, which will be presented in the following pages. However, the feasibility of obtaining sufficient numbers of these ‘expanded SPF’ animals to support toxicology studies is questionable. During the session, general and specific questions and answers were reviewed to guide the discussion and information shared.
Are ‘cleaner’ non-human primates warranted for safety assessment studies in support of clinical trials?
In general, it is feasible to have monkeys with fewer background infections; however, it would be impossible to completely eliminate all background organisms that have the potential to cause overt infectious disease under the appropriate circumstances. The availability of pre-screening or pre-treatment paradigms for NHP depends on the organism of concern.
A major concern for many companies working on molecules targeting specific areas of the immune system is an occasional finding indicating an increased incidence of an infection in a test article-treated group during the course of a toxicology study. In the case of an immune modulator, it may be that the increased incidence of infection results from immune suppression, but the actual morbidity or histopathology may only be attributable to the infection and not the test article. How can one differentiate infection pathology versus test article-related pathology? The first step is to understand the molecule and the resulting manipulation of the immune system. The second step is to understand the immunobiology of a particular infection that appears; for many organisms, this may not be known or fully understood. Follow-up steps would include considerations for pre-screening and/or pre-treating identified pathogens in monkeys for use on future studies to possibly better define a mechanism. This is not an easy decision or task, as some infections, especially viral (e.g., SV40, LCV, RRV) infections, have high prevalence in NHP populations and there are currently no effective pre-treatments. In addition, there may not be a reliable pre-screening method.
SPF must be defined
Importantly, SPF must be defined/qualified to impart meaning (Morton et al., Citation2008). The definition should include a list of pathogens that are desired to be absent from the NHP, which implies ongoing testing and surveillance to verify the absence of specific pathogens in a given population. Some common definitions of SPF macaques are below [see references NCRR-NIH (Citation2009), and Shelburne and Hamill (Citation2003)]:
SPF-1 = Type D simian retrovirus (SRV)-free
SPF-2 = Type D SRV- and simian T-cell leukemia virus (STLV)-free
SPF-3 = SRV, STLV, simian immunodeficiency virus (SIV) or Cercopithecine herpes virus 1 (CHV-1; Herpes B-virus), but must be explicit about SIV or B-virus, which can vary depending on facility
SPF-4 (Level 1) = macaques derived from conventional colony (founders), serologically negative for SRV, STLV, SIV, and B-virus
SPF-4 (Level 2) = macaques born in a colony (usually offspring of founders) free of SRV, STLV, SIV, and B-virus
In addition to viral SPF criteria, Mycobacterium tuberculosis (TB) screening is always a routine event. TB can be introduced into a group from a latently-infected animal or an infected animal handler. Because M. tuberculosis can rapidly spread by the aerosol route, and is zoonotic, it is on every list of pathogens defining SPF macaque colonies (Mansfield, Citation2005; Lerche and Simmons, Citation2008; Morton et al., Citation2008).
Currently, there is no practical way of controlling the definition of SPF, although efforts have been made for some important viruses, as mentioned in the definitions above, based upon safety to human laboratory animal facility workers. In addition to the SPF definitions, there can be other groups defined as ‘super-clean’ or ‘super-SPF’ and would have NHP free of 9-13 pathogens (SPF-9 or SPF-13). Overall, it was agreed that it is important for the NHP user to define the level of SPF status that best suits the intentions of the study design.
Pre-screening and pre-treatment activities
An important variable that can dramatically affect interpretation of toxicity evaluations in NHP is an individual animal’s susceptibility to specific pathogens as well as the type(s) of pathogens it harbors. This pathogen burden is directly related to regional animal origin/source environments, husbandry, and welfare. Pre-screening may be used to eliminate animals from a study prior to dosing so that the remaining animals are as free as possible from the screened-for organism(s). Alternatively, pre-screening may help identify confounding variables so that infection status prior to dosing is at least understood and taken into account for data interpretation. Currently, the industry is moving toward more multiplex technologies, so screening NHP for more infections at once with less serum is possible. Treating animals for specific background infections prior to study start or vaccination against pathogens that may be introduced into the animal colony during a study may help control for some of the variables and minimize ‘noise’ in data generated on the study. However, as mentioned repeatedly during the session, pre-treatment should be balanced against the potential for missing an opportunity to demonstrate increased risk of infection by making animals too ‘clean’.
Viruses
‘It depends’ is part of the best answer to whether one should test animals that have viruses more closely related to human infections. These animals may be more representative, but there are also exceptions with some concurrent NHP viral infections, such as SRV, as there is no known human equivalent. Some relevant questions for consideration include, should one exclude infected NHP for chronic studies? Will there be long-term effects of a test article that may influence the course of an infection? One example included a macaque infected with lymphocryptovirus (LCV) that had no signs/symptoms early on, but later had a clinical presentation similar to Epstein–Barr virus (EBV) in humans. Which viruses are acceptable or desirable as background infections, and which are of more concern? Retroviruses are not as prevalent in many NHP colonies as are certain herpes viruses. Without pre-screening, assignment to dose groups could be unbalanced and confound interpretation. However, some of the viruses that can impact upon toxicology studies have a high rate of false negative results in animals in a screening paradigm.
It was not recommended to assume certain sourcing backgrounds of animals are free of any infections based on historical records. One should be aware of the source of animals and background infections in that colony within a recent time period. Most attendees agreed if sourcing background indicates 98% cytomegalovirus (CMV) positivity, there is no need to screen individuals. However, it is best to identify pre-study serologic status for herpes viruses in general. It is important to remember that the most common age for viral transmission occurs from 6 months to 2 years of age, which encompasses most of the NHP used on toxicology studies. LCV (mentioned earlier and discussed in more detail in the following sessions) and other viruses can remain asymptomatic in immunocompetent NHP, but cause clinical disease during the course of a study due to immunosuppression by the test article or toxicity-related stress, although in most cases asymptomatic NHP will be seropositive.
Bacteria
In considering screening for bacteria, there was expressed concern regarding one case with a non-TB mycobacterium, M. avium, and an immunosuppressant resulting in the depletion of CD4+ T-lymphocytes. However, M. avium infections were explained as pervasive, and considered a ‘truly subclinical opportunistic pathogen’ that exists in soil, food, and water, with disease manifestations only seen in immunocompromised animals. Disseminated M. avium is rare, and it is not possible to eliminate completely from environment.
Attendees indicated that enteric bacteria (i.e., gut flora) can be very fastidious and difficult to culture. Shigella needs to be cultured for three consecutive days and only after three tests can one consider an animal truly negative. Additionally, many bacteria die when exposed to air so there is an initiative to develop PCR methods. One example was mentioned in which 118 animals of three different NHP species were screened and shown to be positive for Shigella, but were all clinically normal, which is unusual and exemplifies what is more commonly observed with Campylobacter infection. Finding Shigella is significant and it should be eliminated. However, one must keep in mind that an organism cultured from an animal with a clinical condition such as diarrhea is not necessarily the cause of the condition, and in an asymptomatic individual, identification of a potential pathogen does not necessarily mean clinical disease will occur during the course of a toxicology study. Importantly, in pre-treating for bacterial infections, inappropriate use of antibiotics in NHP can lead to resistance and care should be taken.
Parasites
NHP are routinely treated for common parasites including tapeworms, whipworms, hookworms, and some protozoa such as Giardia sp. Attendees agreed that parasite-free status was not possible for most parasites, an exception being Plasmodium sp. Most indicated they would pre-treat for Plasmodium, but indicated this type of infection was not as prevalent or of as great a concern as viral infections. However, Plasmodium can have very profound effects on red blood cell (RBC) parameters (e.g., RBC count, hemoglobin, and hematocrit) and specific organs, such as the liver and spleen when recrudescence (i.e., active infection from a dormant or undetectable state) occurs.
The incidence of infection and species of Plasmodium varies considerably depending upon the source of NHP. It is possible to pre-screen for Plasmodium using classical ‘gold standard’ methods of thick and, to a lesser extent thin, blood smears to count parasites in the blood, which can be very time consuming and requires training. More recently, PCR assays have been developed that are very sensitive (0.5 parasites/µL blood) and can determine species of Plasmodium in addition to determining levels of parasitemia prior to and during the course of a study. Additionally, it is possible to safely and effectively treat NHP for Plasmodium infections prior to placement on a study with courses of chloroquine, primaquine, and/or mefloquine.
Colonies of NHP that are largely pathogen/disease-free can be derived by caesarian-derivation and isolation practices, depending on the pathogens, but these practices must be balanced against animal welfare considerations. Some agents are difficult to eradicate regardless of concerted efforts. However, monkeys have been routinely infected and cured with other agents, such as Plasmodium, in searching for effective anti-malarial therapies, with no overt detrimental effects to the animals.
In conclusion, some pathogens should be eliminated for the welfare of the animals on study or safety of the people working with the animals (SRV, Herpes B, and TB); however, in general, knowledge for interpretation of toxicological data, rather than the use of cleaner monkeys is to the greater benefit of human risk assessment and hazard identification.
Expert sources and references regarding NHP infections include attendees of this meeting, members of the Immunotoxicology Technical Committee, ILSI-HESI, and the Committee for the Evaluation of Pre-screening Activities in Non-human Primates (CEPAN). CEPAN was formed during the 2007 Society of Toxicology annual meeting, based on need in the biopharmaceutical industry, and specifically to address the potential for pre-screening/pre-treating NHP infections in response to infections incurred during general toxicology studies used to inform safety assessment in human patients. CEPAN currently exists as a body of 25 experts in the field of NHP infections, care and use.
Discussion Session 2: Treatment, interpretation, and implications of infections occurring during toxicology studies
Karen Price, Katrina Taylor, and Laine Peyton Myers
Toxicity studies of immunomodulatory compounds or drug candidates can present with special considerations that other compounds may not, such as the potential for non-dose-dependent secondary bacterial infections. This discussion section reviewed considerations related to diagnosis and treatment of monkeys within toxicity studies that present with clinical symptomology consistent with secondary infections and regulatory considerations/implications. Overall, attendees agreed animal welfare could not be compromised.
Is it appropriate to treat monkeys that develop infections on toxicity studies or is there danger of masking an immunosuppressive effect?
During the course of a toxicity study, if monkeys present with clinical signs consistent with infection or infestation such as abscesses, purulent discharge, diarrhea, epistaxis, coughing, etc., treatment should be considered, usually within 3 hr of clinical signs identification. Prior to implementing an anti-infective treatment, other causes of potential toxicity should be ruled out. The collection and analysis of specimens to help to diagnose the condition, and assess the animal’s overall health status is recommended. These may include aerobic and anaerobic cultures of blood, discharges, rectum or tissues, fecal flotation, testing for clostridium toxin, examination of feces, complete hematology and serum chemistry analysis, or other diagnostic measures consistent with the presenting symptoms. Based on the results, the most appropriate treatment options can then be evaluated in consultation with a clinical veterinarian. In addition, if a causative agent is identified, knowledge of whether that agent is a known pathogen or opportunistic infection may be valuable for determining potential test article-related immunosuppression.
Common enteral pathogens include Campylobacter jejuni, Campylobacter coli, enteropathogenic Escherichia coli, Salmonella spp., Shigella spp., Helicobacter spp., and Yersinia spp. These organisms present with varying severity of diarrhea and may be accompanied by bloody or mucoid stools, vomiting, and/or wasting. Commonly-used treatments include supportive care, enrofloxacin, ampicillin, amoxicillin, and metronidazole. Treatment for some species, such as Salmonella spp., may be a concern as animals may become carriers, and multidrug resistance can be common (Fox et al., Citation2002; Taylor, Citation2010).
Common respiratory pathogens in monkeys include Streptococcus pneumoniae, Klebsiella pneumoniae, and Moraxella catarrhalis (VandeWoude and Luzarraga, Citation1991; Fox et al., Citation2002). Presentation may consist of epistaxis (VandeWoude and Luzarraga, Citation1991), upper respiratory disease (nasal discharge, congestion), meningitis, cough, and/or dyspnea, and the infections may be treated with enrofloxacin, penicillin, and/or cephalosporins (Fox et al., Citation2002; Taylor, Citation2010). Other bacterial infections such as skin abscesses secondary to trauma are commonly caused by Staphylococcus and Streptococcus spp. Common treatments include penicillin, amoxicillin, enrofloxacin, and cephalosporins [for a more detailed review of these pathogens commonly observed in monkeys, see Taylor (Citation2010) in this issue].
Within the context of a toxicity study, a tiered approach to diagnosis and treatment may be warranted. First and foremost, animal welfare should be the primary concern, and a qualified veterinarian should thoroughly examine the animals and provide guidance on their care. Gastrointestinal disturbances are quite common in outbred monkey colonies and can lead to rapid debilitation and mortality in rhesus macaques (Hird et al., Citation1984; George and Lerche, Citation1990; Elmore et al., Citation1992; Sestak et al., Citation2003). This is complicated by the low reserve capacity of body weight in several monkey species, particularly the more commonly-used cynomolgus macaques, which may be only 2–5 kg in weight. Thus, the timely recognition of symptoms by the technical staff and initiation of supportive measures is critical and allows time for further diagnostic work-up. The manner in which the culture is collected is crucial, especially for fecal cultures, and may not always be readily obtained from sick animals. If the toxicity profile of the compound is not yet known and/or the compound is in the initial stages of testing, it may be acceptable to let the suspected infection or clinical symptomology progress with supportive care (fluid replacement, nutrition, etc.) for the animal as appropriate, until euthanasia is necessary for animal welfare reasons. This option provides for a complete clinical and histopathological workup of the animal that may determine the cause of illness/death and help distinguish between any primary target-organ toxicities and diagnosis of infection secondary to drug-related immunomodulation. Once an infection has been confirmed as cause of toxicity in the animal, it may then be appropriate to use more aggressive measures of treatment or intervention for future occurrences. These measures may include antibiotic treatment, dose lowering of the immunosuppressive agent, or temporary withdrawal from drug treatment (drug holiday). Depending on the type of infection, pain medication may also be provided if the animal is noticeably uncomfortable. By using one or more of these measures, the monkey may be able to recover from the infection and recommence dosing for the remainder of the study. This may be especially helpful in longer-term studies where it is crucial to maintain a powered study for the intended duration of dosing to identify other primary or unintended effects of the test compound.
When determining the course of action to treat a suspected infection on a toxicology study, it is important to know and document the toxicological profile of the medications that are being considered. Some conditions may require euthanasia in place of intervention or even after intervention has been initiated. To balance this, it is useful to ensure that the study is powered with a statistically-significant number of animals.
Most attendees agreed that although it is preferable to limit additional variables during non-clinical studies, antibiotic treatment of animals on study need not be avoided. When possible, a diagnostic and treatment plan for response to infection in monkey toxicity studies of immunosuppressive agents should be considered when writing the non-clinical protocol.
Early detection and diagnostics are vital to the potential of controlling or eliminating secondary infection. As discussed earlier, close monitoring of monkeys is needed. In non-human primates (NHP), the presentation of clinical signs of infection often occurs when the infection is already substantial or spreading further. Animals with active lesions, diarrhea or significant histories of injuries or diarrhea may not be appropriate for use on studies of immunosuppressive test compounds. If selection is necessary, treatment of the condition prior to the study may be performed and documented.
What additional data are needed to aid in the interpretation of toxicology findings in monkeys that may be related to infection?
If possible, diagnostic testing such as aerobic and anaerobic cultures of blood, discharges, rectum or tissues, fecal flotation, clostridium toxin examination of feces, complete blood count or other diagnostic measure at the time of symptom recognition should be collected and performed. Provisions should also be made to collect diagnostic samples at necropsy for direct correlation with histopathology.
Knowledge of pre-study and stock colony health rates of infection within and across study facilities is imperative when determining whether an infection that occurs on study is incidental or is at increased propensity in drug-treated groups because of drug-related immunosuppression. If a bacterial or parasitic pathogen (i.e., Shigella, Giardia) is detected in a stock monkey colony prior to a study with an immunomodulatory agent, treatment of the entire colony may be considered. Due to the variation in viral status of monkeys from different sources, the source of monkeys used for a particular drug program should be kept as consistent as possible to minimize confounding results across studies and maintain consistency in terms of viral status.
Knowledge of the pharmacologic target and its interactions with the immune system can aid in determining a drug relationship and ultimately, the diagnosis of infection. Using specific functional assays, demonstration of a direct drug effect on a particular immune cell type or immune mechanism at exposures of drug associated with infections in the animals may help determine a drug relationship (Price, Citation2010). Using a tiered approach, functional assays may provide additional information following the initial routine screening to measure one or more components of the total immunological response. These may include non-specific or innate cell endpoints such as natural killer (NK) cell activity, neutrophil and monocyte/macrophage function (phagocytosis and respiratory burst), and evaluation of specific acquired immunity including T-dependent assays to quantify antibody responses to neoantigens such as keyhole limpet hemocyanin, sheep red blood cells, ovalbumin, tetanus toxoid, or influenza virus to evaluate humoral immunity, and assays to measure cell-mediated immunity such as cytotoxic T-lymphocyte assays or those to measure delayed-type hypersensitivity (DTH). These specialized assays/endpoints may be assessed ex vivo on toxicity studies and can provide evidence of potential for reversibility (Price, Citation2010) even when recovery from an infection was not entirely demonstrated within the context of a toxicology study and given a possible lack of reproducibility of this infection within the toxicology program. In some cases, these or similar endpoints may be used as biomarkers in clinical studies.
How do we interpret the outcome of animals that are treated for infection versus those which are not? How should data be interpreted within a study and across studies with the same drug where infections are seen in some studies and not in others?
As with all anti-infective medications, individual animal response may vary, and side effects of the medications used may vary. Treatment of all animals in a study even if they are not presenting with clinical signs or have diagnostic evidence of infection should be considered only in extreme cases. It should be noted that in studies with a low (underpowered) number of animals, background rates of infection may be difficult to establish, again demonstrating the need for an adequately powered study.
Latent infections of plasmodia, retroviral diseases, and/or Shigella/Campylobacter can also be subclinical and difficult to detect until an immunosuppressive agent is on board. As discussed earlier, knowledge of the origin of animals, their clinical history, and thorough pre-test work-ups are essential for selection of animals assigned to the study as well as interpretation of infection incidence within a study and across studies in a given program.
What are the regulatory implications when infection is observed as an expected pharmacologic effect?
The occurrence of infections in non-clinical monkey toxicology studies of immunosuppressive drugs, when not properly diagnosed, can confound the interpretation of study results. Because monkey toxicology studies are used to assess human risk, it is important to correctly diagnose the infection and, subsequently, to recognize the differences between monkeys and humans in terms of their susceptibility and capacity to deal with certain types of infections and related symptomology. Thus, a detailed documentation strategy can be helpful and may include a comprehensive testing and treatment plan, comparison of results between dose levels and across studies, and investigative mechanism of action studies to address the immunotoxicologic mechanism, which can help characterize the effects observed and lead to a proper diagnosis.
Because human risk assessment is often the ultimate goal of monkey toxicology studies, the clinical indication of the compound also needs to be considered when determining the potential clinical implication of infections observed during the non-clinical studies. Depending on the types of infections observed and their relevance to humans, there may be a need to proceed more cautiously or implement strict stopping criteria in certain indications. In addition, differences between NHP and humans can dictate side effects and secondary infection rate differences in preclinical versus clinical trials. Key differences include the low reserve capacity of several monkey species, as previously discussed, which magnifies and accelerates the consequences of inappetence and reduced water consumption, and relatively poor hygiene (i.e., lack of daily cleansing, hand washing, etc.) and possibility for fecal-oral and cage-to-cage transmission of bacteria (Taylor, Citation2010). This may result in rapid progression of infection such that traditional intervention techniques will not be beneficial. Some infections or infestations may be relevant in one species but not in the other.
FDA Reviewers concerns include, not only the infection, but addressing questions such as ‘what are the pharmacologic or toxicologic effects of the additional treatment (the anti-microbial) that was added to the study? Does this confound interpretation? Are the findings dose-responsive? Is the mechanism of action of the immunosuppression relevant to humans (i.e., will the suppression likely occur in humans at the same rate or are humans less sensitive to the response associated with a particular receptor/pathway?)’.
Discussion Session 3: Viral status can indicate immunosuppression in non-human primates
David Hutto and Yanli Ouyang
There are a few viral diseases that have been shown to occur only in or at a higher rate in immunosuppressed humans and non-human primates (NHP), so-called opportunistic viral infections. These include diseases related to BK virus, Epstein-Barr Virus (EBV), human parvovirus B19, and cytomegalovirus (CMV) in humans, and their respective simian homologues SV40, lymphocryptovirus (LCV), simian parvovirus, and simian CMV, respectively. These diseases in NHP have been described primarily in relationship to the immunosuppressive lentiviruses, simian immunodeficiency virus (SIV), and the chimeric simian/human immunodeficiency virus or following the use of immunosuppressive solid organ transplantation regimes. As these viruses generally cause disease only under conditions of immunomodulation, we have given due consideration to the proposal that monitoring the viral status of NHP exposed to immunomodulatory therapeutics during non-clinical safety assessment studies could serve as an indirect measure of the immune reserve in these animals.
Viruses of interest
The utility of quantitatively assessing viral nucleic acid burden in peripheral blood (viral load) has been established in several human and NHP clinical settings. This is done by quantitative polymerase chain reaction (qPCR) methods, and is often coupled with ancillary serologic assessment to provide a more complete picture of viral status coupled with specific humoral immune response to the virus under study.
Viral load has been established as a sensitive predictor of future and current disease status in infections with HIV and SIV. In both humans and NHP, viral load of EBV and LCV, respectively, has been shown to have both positive and negative predictive power with regard to progression to post-transplant lymphoproliferative disease under conditions of immunosuppression intentionally induced for allograft transplantation procedures. Similarly, viral load for the polyomavirus BK virus has shown correlation with severity of renal allograft rejection. CMV viral load has also been shown to be predictive of CMV disease in human liver transplants and in SIV infected rhesus macaques (Mellors et al., Citation1996; Watson et al., Citation1997; Humar et al., Citation1999; Limaye et al., Citation2001; Wagner et al., Citation2001; Rivailler et al., Citation2004).
There is evidence, then, that viral load for these agents is correlated with and is predictive of the degree of immunosuppression in the host. Indeed, in the clinical transplant setting these viral load data have been successfully used to support back titration of the immunosuppressive regimen in order to decrease viral load, thereby diminishing the chance for onset of viral disease and concurrently maintaining dampening of the host immune response against the graft organ. A unifying theme for all of these specific examples relating viral load to immunosuppression is a decrease in the immune reserve specifically mediated by T-lymphocytes. The lentiviruses are clearly T-lymphocyte depleters and most of the drugs used to induce immunosuppression for the purposes of transplantation, such as cyclosporine A, are uniquely effective against T-lymphocyte-mediated immune responses (Granelli-Piperno et al., Citation1988; Ho et al., Citation1996; McCune, Citation2001; Grossman et al., Citation2002).
Applications to safety assessment of immunomodulatory pharmaceuticals
There appear to be two general scenarios when monitoring viral load in NHP could be of potential value in the non-clinical safety assessment of immunomodulatory pharmaceuticals. The first scenario could come into play if a drug targets a key component of an immunologic pathway, the interruption of which is known to be associated with specific viral opportunistic infections or viral recrudescences, such as T-lymphocyte-mediated immunity. In such a situation it may be prudent to build into non-clinical safety assessment studies either evaluation (or collection) of the appropriate samples for subsequent evaluation of the viral load for the predicted agent. A second scenario is where unusual or unexpected findings related to immunomodulation occur in a non-clinical safety assessment study and there is a desire to retrospectively assess if viral load could be a useful monitoring tool with regards to immune reserve in future non-clinical or clinical studies with that immunomodulatory drug. Given that it is difficult to prospectively identify such circumstances and given that special sample collection procedures must be followed to facilitate viral load assessments, this second scenario is probably the more common of the two, but the least likely to be successfully pursued.
Challenges associated with monitoring viral load in non-clinical safety assessment studies are not trivial. As noted, most demonstrated associations between immunomodulation and increases in the viral load of recrudescent or opportunistic viruses are associated with T-lymphocyte modulation. It would seem logical that this effect on viral load might also occur in NHP exposed to immunomodulators that target non-T-lymphocyte components of the immune system, but data supporting that assumption are not available. Thus, in the face of a lack of effects on viral load induced by a non-T-lymphocyte immunomodulatory drug, there is no historical data to support a conclusion that the drug does not negatively impact immune reserve, as measured by that endpoint.
Another challenge with monitoring viral load is that sample collection requirements for qPCR samples are outside the range of what is normally entailed in a routine toxicity study for pharmaceuticals. Since the most likely scenario is that viral load will be of interest after a study has been completed, and the necessary samples will most likely not have been collected, the only apparent solution is to include collection of qPCR appropriate samples routinely in all non-clinical safety studies. However, until there is clear evidence that interdiction in immune targets that involve non-T-lymphocyte immune components results in measurable alterations in viral load, planning for routine inclusion of this endpoint in all toxicity studies supporting immunomodulatory drugs cannot be recommended.
There are numerous other measures of immune function and reserve that can be applied in toxicity studies with NHP. They are not the topic of discussion here but are noted for purposes of completeness and include: (i) delayed-type hypersensitivity (DTH) response; (ii) innate immune system evaluations (neutrophil and macrophage assays, levels of complement components, antibody-dependent cell-mediated cytotoxicity and NK assays); and (iii) novel, untested measures (ELISPOT assays for antigen-specific antibody-secreting cell enumeration, ex vivo cytokine release assays, etc).
Potential impact of viral load data
When new technologies and collectable endpoints present themselves as being applicable to toxicity studies, the immediate considerations that arise are whether the endpoint is relevant to humans and how the collected data will potentially affect clinical use of the drug. Regarding viral load specifically, two such questions are: (i) is an increase in viral load alone adverse (in the absence of correlative clinical or pathology data)? and, (ii) should clinical doses in humans be affected by changes in viral load? Without more historical or prospectively-collected experimental data, these are difficult questions to answer. Therefore, if there is interest in using viral load as a useful indicator of immune status in non-clinical safety assessment studies, qualification of the endpoint is required. This qualification effort would require dedicated studies with NHP exposed to established control immunomodulatory test articles that impact different aspects of immunity that are known to impact viral load positively and negatively, respectively. It seems unlikely at this time that any pharmaceutical company, contract research organization, or government agency will be willing to assume responsibility for this qualification effort. Without qualification, viral load as an endpoint in NHP toxicity studies cannot be considered to have positive or negative predictivity with regards to degree of immunomodulation, immune reserve, or human relevance of the finding in NHP.
Conclusion
These discussions would indicate the need for an objective and informed study design-dependent approach to determining the need for and degree of SPF status appropriate for toxicology studies, identifying pathogens for pre-screening or pre-treatment, interpretation implications and paradigms for treating specific NHP pathogens during a toxicology study, and the utility of infections for immune surveillance. For the most informed path forward, it is important to understand the interaction between the infectious organism and the host immune system, as is discussed in the preceeding publications of presentations during the morning sessions (Hutto, Citation2010; Lerche, Citation2010; Price, Citation2010; Sasseville and Mansfield, Citation2010; Simmons, Citation2010; Taylor, Citation2010).
Based on Discussion session 1, defining the SPF status of NHP to be used on toxicology studies will ensure those pathogens of most concern that may confound finding interpretation are considered measurably absent through available pre-screening methods. Pre-treatment for certain pathogens can further prevent unwanted or unanticipated perturbations in standard toxicological endpoints. If there are persistent or untreatable infections, pre-screening information would certainly inform selection and distribution of NHP for a study, and enable an awareness of potential perturbations in data obtained. The use of cleaner monkeys may not predict potential unanticipated immunotoxicities however. Unless all monkeys were infected with the same pathogens at the same level, which would be biologically impossible, it would be difficult to attribute pathological changes to the intended single variable.
Based on Discussion session 2, the origin of the animals being used, the history of each individual animal, thorough pre-test screening of the animals, background incidence of disease in the colony, the number and distribution of animals that are affected, any diagnostic testing that was performed, the treatments that were necessary and histopathology are vital information to correlate infections that occur in non-clinical toxicology studies with immunosuppression versus direct compound effects. A thorough work-up of any animal that presents with concerns of effects potentially related to infection should be performed. Once a diagnosis is made, conservative treatment should be promptly initiated. Provided there are no concerns for drug/drug interactions or potential effects on the metabolism of the test article, it is acceptable to use anti-infectives on toxicology studies, when deemed appropriate by the clinical veterinarian. If the animal does not respond to appropriate treatment, a short drug (test article) holiday (if feasible, i.e., for small molecule compounds) may be considered in longer-term studies until the animal shows signs of recovery. For biologicals, the half-life must be considered when determining the feasibility of drug interruption. If the humane condition of the animal is compromised, then euthanasia should be considered. In monkeys treated with immunosuppressive agents, the occurrence of infection may be considered a secondary pharmacologic effect. Infections that resolve despite continued dosing may not be considered adverse depending on the agent and/or the symptomology of the infection. In addition, the lack of a no-effect level in a single-study due solely to the presence of infections may not necessitate repeating the study or delaying clinical trials if a suitable number of animals are unaffected or are able to resolve the infection with appropriate intervention. However, it is essential to have a strict monitoring plan in place for infections in clinical studies with associated inclusion/exclusion, therapeutic, and stopping criteria. This will aid in ameliorating regulatory issues that may arise when submitting the study results.
Based on Discussion session 3, assessment of viral load as a measure of risk for development of disease in humans and NHP has established clinical utility in certain very specific instances involving well-characterized viruses in well-understood clinical situations. This endpoint is of unknown general predictive value in regards to assessing general immune modulation or immune reserve in NHP exposed to a broad range of immunomodulatory therapeutics. Thus, routine evaluation of this endpoint is not recommended in initial NHP safety assessment studies of novel immunomodulatory therapeutics.
Acknowledgements
The authors express their most sincere and immense gratitude to Raegan O’Lone, without whom this extensive collection of manuscripts would not have been possible.
Declaration of interest
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References
- Elmore, D. B., Anderson, J. H., Hird, D. W., Sanders, K. D., and Lerche, N. W. 1992. Diarrhea rates and risk factors for developing chronic diarrhea in infant and juvenile rhesus monkeys. Lab. Anim. Sci. 42:356–359.
- Fox, J. G., Anderson, L. A., Loew, F. M., and Quimby, F. W. (Eds.). 2002. Laboratory Animal Medicine, 2nd Edition. New York: Academic Press, pp. 676–791.
- George, J. W., and Lerche, N. W. 1990. Electrolyte abnormalities associated with diarrhea in rhesus monkeys: 100 cases (1986-1987). J. Am. Vet. Med. Assoc. 196:1654–1658.
- Granelli-Piperno, A., Keane, M., and Steinman, R. M. 1988. Evidence that cyclosporine inhibits cell-medicated immunity primarily at the level of the T-lymphocyte rather than the accessory cell. Transplantation 46: 53S–60S.
- Grossman, Z., Meier-Schellersheim M., Sousa, A., Victorino, R., and Paul, W. 2002. CD4+ T-cell depletion in HIV infection: Are we closer to understanding the cause? Nature Med. 8:319–323.
- Hird, D. W., Anderson, J. H., and Bielitzki, J. T. 1984 Diarrhea in non-human primates: A survey of primate colonies for incidence rates and clinical opinion. Lab. Anim. Sci. 34:465–470.
- Ho, S., Clipstone, N., Timmermann, L., Northrop, J., Graef, I., Fiorentino, D., Nourse, J., and Crabtree, G. R. 1996. The mechanism of action of cyclosporine A and FK506. Clin. Immunol. Immunopath. 80:S40–S45.
- Humar, A., Gregson, D., Caliendo, A., McGeer, A., Malkan, G., Krajden, M., Corey, P., Greig, P., Walmsley, S., Levy, G., and Mazzulli, T. 1999. Clinical utility of quantitative cytomegalovirus viral load determination for predicting cytomegalovirus disease in liver transplant recipients. Transplantation 68:1305–1311.
- Hutto, D. L. 2010. Opportunistic infections in non-human primates exposed to immunomoduatory biotherapeutics: Considerations and case examples. J. Immunotoxicol. (In Press).
- Lerche, N. W. 2010. Simian retroviruses: Infection and disease - Implications for immunotoxicology research in primates. J. Immunotoxicol. (In Press).
- Lerche, N. W., and Simmons, J. H. 2008. Beyond specific pathogen-free: Biology and effect of common viruses in macaques. Comp. Med. 58:8–10.
- Limaye, A., Jerome, K. R., Kuhr, C. S., Ferrenberg, J., Huang, M., Davis, C. L., Corey, L. and Marsh, C. L. 2001. Quantitation of BK virus load in serum for the diagnosis of BK virus–associated nephropathy in renal transplant recipients. J. Infect. Dis. 183:1669–1672.
- Mansfield, K. 2005. Development of specific pathogen free non-human primate colonies. In: The Laboratory Primate (Wolfe-Coote, S., Ed.). London: Elsevier Academic Press, pp. 229–239.
- McCune, J.M. 2001. The dynamics of CD4+ T-cell depletion in HIV disease. Nature 410:974–979.
- Mellors, J. W., Rinaldo Jr. C. R., Gupta, P., White, R. M., Todd, J. A., Kingsley, L. A. 1996. Prognosis in HIV-1 infection predicted by the quantity of virus in plasma. Science 24:1167–1170.
- Morton, W. R., Agy, M. B., Capuano, S. V., and Grant, R. F. 2008. Specific pathogen-free macaques: Definition, history and current production. ILAR J. 49:137–144.
- NCRR-NIH. 2009. Non-human Primate Research Resources. National Center for Research Resources Fact Sheet. http://www.ncrr.nih.gov/publications/pdf/NPRR_Fact_Sheet_508.pdf.
- Price, K. D. 2010. Bacterial infections in cynomolgus monkeys given small molecule immunomodulatory antagonists. J. Immunotoxicol. (In Press).
- Rivailler, P., Carville, A., Kaur, A., Rao, P., Quink, C., Kutok, J. L., Westmoreland, S., Klumpp, S., Simon, M., Aster, J. C., and Wang, F. 2004. Experimental rhesus lymphocryptovirus infection in immunosuppressed macaques: An animal model for Epstein-Barr virus pathogenesis in the immunosuppressed host. Blood 104:1482–1489.
- Sasseville, V. G., and Mansfield, K. G. 2010. Overview of known non-human primate pathogens with potential to affect colonies used for toxicity testing. J. Immunotoxicol. (In Press).
- Sestak, K., Merritt, C. K., Borda, J., Saylor, E., Schwamberger, S. R., Cogswell, F., Didier, E. S., Didier, P. J., Plauche, G., Bohm, R. P., Aye, P. P., Alexa, P., Ward, R. L., and Lackner, A. A. 2003. Infectious agent and immune response characteristics of chronic enterocolitis in captive rhesus macaques. Infect. Immun.71:4078–4086.
- Shelburne, S. A., and Hamill, R. J. 2003.The immune reconstitution inflammatory syndrome (IRIS). AIDS Rev. 5:67–79.
- Simmons, J. H. 2010. Herpesvirus infections of laboratory macaques. J. Immunotoxicol. (In Press).
- Taylor, K. 2010. Clinical veterinarian’s perspective of non-human primate (NHP) use in drug safety studies. J. Immunotoxicol. (In Press).
- VandeWoude, S. J., and Luzarraga, M. B. 1991. The role of Branhamella catarrhalis in the “bloody-nose syndrome” of cynomolgus macaques. Lab. Anim. Sci. 41:401–406.
- Vickers, J. H. 1969. Diseases of primates affecting the choice of species for toxicologic studies. Ann. N. Y. Acad. Sci. 162:659–672.
- Wagner, H. J., Wessel, M., Jabs, W., Smets, F., Fischer, L., Offner, G., and Bucsky, P.2001. Patients at risk for development of post-transplant lymphoproliferative disorder: Plasma versus peripheral blood mononuclear cells as material for quantification of Epstein-Barr viral load by using real-time quantitative polymerase chain reaction. Transplantation 72:1012–1019.
- Watson, A., Ranchalis, J., Travis, B., McClure, J., Sutton, W., Johnson N, P. R., Hu S, L., and Haigwood, N. L. 1997. Plasma viremia in macaques infected with simian immunodeficiency virus: Plasma viral load early in infection predicts survival. J. Virol. 71:284–290.