Nonclinical Safety Considerations for the Development of Pediatric-First Drugs: An Industry View.

This paper provides considerations on approaches to the development of medicines initially developed for pediatric use (ie, "pediatric-first" or "pediatric-only" drugs). The most common development approach for these types of medicines involves a first-in-human (FIH) clinical trial with healthy adult volunteers to assess safety and tolerability. This approach generally requires nonclinical repeat-dose studies in adult animals; safety pharmacology and in vivo genetic toxicology studies in adult animals are also performed for small-molecule drugs. Additional studies in juvenile animals may be required prior to clinical trials in pediatric patients, on a case-by-case basis. In this paradigm, the starting dose for pediatric patients is primarily driven by modeling from the adult pharmacokinetic assessment and pharmacology data. A second development approach is where the FIH clinical trial is conducted in pediatric patients. This approach is generally supported by repeat-dose studies in juvenile animals, with the onset of dosing at ages that developmentally correlate to the age of the pediatric patients. Safety pharmacology and in vivo genetic toxicology studies are generally performed in adult animals for small-molecule drugs. To define a safe yet minimally efficacious starting dose for pediatric patients, various complementing approaches can be used, including human equivalent dose, minimal anticipated biological effect level, and physiologically based pharmacokinetic modeling. Case examples for pediatric-first drug candidates show how both drug development approaches (ie, entry into human first in adults or directly in pediatric populations) are used in the pharmaceutical industry.

The Fc receptor for IgG (Fc gamma RII; CD32) on human neonatal B lymphocytes.

B cells express an Fc receptor for IgG (FcgammaRII; CD32) which is involved in feedback inhibition of antibody production. Engagement of FcgammaRII during ligation of the antigen receptor provides an inhibitory signal. FcgammaRII exists as several isoforms, with FcgammaRIIb (which carries an immunoreceptor tyrosine-based inhibition motif; ITIM) being predominant form on adult B cells. The inhibitory role of FcgammaRIIb may be unhelpful to the infant, since primary exposure to infectious agents is likely to be in the presence of maternal IgG. We hypothesized that neonatal B cells would be less susceptible to feedback inhibition by antibody, either through the expression of activation-competent FcgammaRII isoforms (FcgammaRIIa and FcgammaRIIc) or through reduced expression of the inhibitory FcgammaRIIb isoforms. Cord and adult B cells were examined for expression of FcgammaRII isoforms using monoclonal antibodies and RT-PCR. In vitro assays were performed to assess susceptibility of cord and adult cells to FcgammaRII-mediated suppression. Although there is no phenotypic difference in FcgammaRII expression (FcgammaRIIb predominating on both adult and cord B cells), FcgammaRIIb is expressed at lower levels on cord cells. This quantitative difference in FcgammaRIIb expression may explain the reduced susceptibility of cord B cells to antibody-mediated inhibition observed in these experiments.

Interleukin 2 receptor regulation and IL-2 function in the human infant.

IL-2 receptor is expressed at low levels on adult blood lymphocytes, and at lower levels on cord blood cells. IL-2 receptor alpha and beta chain expression increases gradually from 0-18 months of age. The level of soluble CD25 (IL-2 receptor alpha chain) has been reported to be elevated in cord blood. Quantitative RT-PCR showed that adult cells express 10 times as much CD25 mRNA as cord cells. Cord plasma showed only a marginal ability to strip CD25 from the membrane. To assess the functional consequences of low IL-2 receptor expression, cord and adult cells were activated in vitro. The response was stimulus-dependent, but cord cells upregulated CD25 readily. Cord and adult cells proliferated in an IL-2-dependent assay to a similar extent. Infants suffering acute infection showed marginally higher levels of membrane CD25 expression than infants without overt infection. Thus neonatal and infant lymphocytes express lower levels of IL-2 receptors than adult cells, reflecting lower mRNA concentrations at least for CD25; they are able to up-regulate receptors in response to in vitro stimulation and are able to respond in vitro to IL-2-dependent stimulation; however in vivo there may be a dampening down of the IL-2 system in infancy.

Somatic mutation of immunoglobulin V(H)6 genes in human infants.

Infants respond to antigen by making antibody that is generally of low affinity for antigen. Somatic hypermutation of immunoglobulin genes, and selection of cells expressing mutations with improved affinity for antigen, are the molecular and cellular processes underlying the maturation of antibody affinity. We have reported previously that neonates and infants up to 2 months of age, including individuals undergoing strong immunological challenge, show very few mutated V(H)6 sequences, with low mutation frequencies in mutated sequences, and little evidence of selection. We have now examined immunoglobulin genes from healthy infants between 2 and 10 months old for mutation and evidence of selection. In this age group, the proportion of V(H)6 sequences which are mutated and the mutation frequency in mutated sequences increase with age. There is evidence of selection from 6 months old. These results indicate that the process of affinity maturation, which depends on cognate T-B cell interaction and functional germinal centres, is approaching maturity from 6 months old.

Heterogeneity of CD44 expression among human B-cell subpopulations.

CD44 is a widely distributed cell surface glycoprotein that participates in a number of cellular adhesion and signal transduction processes. Germinal center B cells express very low levels of CD44, whereas their precursors and differentiation products express much higher levels. In immunofluorescence studies comparing 20 antibodies classified as being against the hematopoietic isoform of CD44, one antibody, A1G3, was unreactive with germinal center B cells, whereas the other antibodies showed low intensity but definite reactivity. Western blotting and sequential immunoprecipitation studies of lysates from density-separated lymphocyte fractions showed two bands that were differentially expressed and reacted differently with A1G3 compared with the other CD44 antibodies. These results suggest that germinal center B cells and non-germinal center B cells express different forms of CD44. Of 21 malignant B-cell populations examined, 5 showed reactivity with a "standard" CD44 reagent and significantly reduced reactivity with A1G3, while one sample showed the opposite pattern and the remainder were positive for both reagents. Of 10 cell lines studied, 5 were differentially stained by A1G3 and a standard CD44 antibody. PCR amplification of reverse transcribed mRNA from sorted human tonsil B-cell subpopulations and Southern blotting showed that B cells express a number of splice isoforms of CD44. These results demonstrate that B cells express multiple forms of CD44; both splice insert isoforms and variants distinguished by A1G3; the form of CD44 expressed depends on B-cell differentiation state.

Somatic hypermutation of immunoglobulin genes in human neonates.

The antibody response in the young infant is limited in several ways; in particular, responses generally are of low affinity and restricted to IgM. This raises the question whether the affinity maturation process, consisting of somatic mutation of immunoglobulin genes coupled with selection of high-affinity variants, is operative in the neonate. Re-arranged V(H)6 genes were amplified by polymerase chain reaction (PCR) from cord blood and from peripheral blood of infants. Heteroduplex analysis detected mutation in only 2/18 cord blood samples, while mutations were seen from about 10 days of age onwards. Cloning and sequencing of mutated neonatal V(H)6 genes showed that mutated sequences contained relatively few mutations (one to three mutations per sequence) compared with published values of about 10 in adult IgM sequences. Selection was not evident in the majority of neonatal samples. Thus mutation can occur in the human neonate, but is minimal and generally not accompanied by selection. The age at which affinity maturation develops effectively is yet to be defined.

The antigen receptor complex on cord B lymphocytes.

The neonatal immune system responds to a restricted range of antigens, producing largely IgM antibody of low affinity. Comparison of the components of the B-cell antigen receptor complex shows significantly elevated membrane levels of IgM in neonatal B cells, compared with adult cells. CD79, which acts as the signal transducer for membrane immunoglobulin, is elevated in parallel with IgM, while IgD is elevated to a lesser degree. CD19, CD21, CD22 and CD81, which are all involved in transmitting activation signals when immunoglobulin is engaged, are not elevated. CD32, which is involved in negative regulation of activation, is present at reduced levels on cord B cells. The elevation of B-cell membrane IgM persists during infancy. Neonatal B cells respond in vitro to interleukin-4 (IL-4) by further elevation of membrane IgM levels. The elevated level of membrane IgM may make neonatal B cells easier to trigger by low concentrations of antigen, but in vitro activation and immunoglobulin modulation experiments did not show significant differences between cord and adult B-cell responses to anti-IgM.

The Fas antigen (CD95) on human lymphoid cells: epitope analysis with ten antibodies.

The expression of CD95 antigen was examined on adult and cord blood lymphocytes using a highly sensitive immunofluorescence/flow cytometric procedure. CD95 was expressed by the majority of circulating blood T cells in adults, and by a smaller proportion of CD4+ and CD8+ T cells in cord blood. The majority of circulating B cells did not react with seven CD95 antibodies, but three antibodies did stain B cells. In tonsil sections, CD95 was expressed throughout the tissue but germinal centres showed generally stronger staining than the surrounding follicular mantle and interfollicular areas. This was confirmed by flow cytometry, which showed expression preferentially on B cells with a germinal centre phenotype. Because different antibodies stained different proportions of B cells, CD95 epitopes were examined by inhibition, additive binding and protease susceptibility studies using a panel of ten CD95 antibodies. B cells apparently reacting selectively with CD95 antibodies were sorted and CD95 mRNA was reverse transcribed to cDNA and analyzed, in order to confirm the presence of CD95 in cells which reacted selectively and to explore the possible existence of CD95 isoforms. The major cDNA band was identical in the two populations. Inhibition of N-glycosylation suggested that the epitopes detected differentially could not be accounted for by differential N-glycosylation.

Purification of cord blood lymphocytes.

When cord blood is separated using standard methods based on Ficoll-Hypaque, the mononuclear fraction is contaminated with erythrocytes and also with nucleated cells that do not express the leucocyte marker CD45. The contamination with CD45-negative cells can exceed 50%, and will interfere with phenotypic, mRNA or functional analysis. A large proportion of these cells are erythrocyte precursors. The contaminating cells may be removed by lysis with hypotonic ammonium chloride; when the cells are required for studies which are adversely affected by ammonium chloride (such as antigen processing), high purity can be attained by two rounds of density separation.

erbB family receptor expression and growth regulation in a newly isolated human breast cancer cell line.

A new human breast cancer cell line (SUM-52PE), originating from a malignant pleural effusion specimen, that can be cultured under serum-free conditions has been isolated. Experiments were conducted to examine the relationship between expression of the erbB family of growth factor receptors and growth regulation in these cells. SUM-52PE cells are epidermal growth factor receptor negative but express single copy levels of erbB-2 protein. Southern blot analysis indicates that the erbB-2 gene is not amplified in these cells. The cells also express mRNA for both erbB-3 and erbB-4. Phosphotyrosine Western blot analysis of membrane protein obtained from SUM-52PE cells indicates the presence of a constitutively tyrosine phosphorylated M(r) 185,000 protein. Immunoprecipitation, using antibodies to erbB-2 or erbB-3, coupled to phosphotyrosine Western blot analysis indicates that both erbB-2 and erbB-3 are constitutively tyrosine phosphorylated in proliferating SUM-52PE cells. Conditioned medium obtained from SUM-52PE cells does not induce tyrosine phosphorylation of p185erbB-2 in a sensitive indicator cell line, suggesting that an erbB-2 activating factor is not secreted by these cells. However, neu differentiation factor/heregulin (NDF/HRG) mRNA is expressed by the cells, and Western blot analysis of SUM-52PE membrane protein revealed the presence of a M(r) 90,000 immunoreactive NDF/HRG protein. Thus, SUM-52PE cells synthesize a membrane bound form of NDF/HRG that may activate erbB-2 and erbB-3 via a juxtacrine mechanism. The addition of exogenous beta-2-NDF/HRG to the culture medium of SUM-52PE cells yields enhanced tyrosine phosphorylation of p185erbB-2/erbB-3 but has only a small stimulatory effect on the proliferation of these cells. By contrast, an erbB-2 monoclonal antibody that binds to the extracellular domain of erbB-2 is potently mitogenic for these cells. SUM-52PE cells were also found, by phosphotyrosine Western blot analysis, to express an inordinately large number of tyrosine phosphoproteins. Direct measurement of phosphotyrosine phosphatase (PTPase) activity in SUM-52PE cell membrane protein revealed 2-3-fold lower levels of PTPase activity compared to other normal and neoplastic breast epithelial cell lines. Thus, SUM-52PE cells exhibit altered growth phenotypes not identified previously in human breast cancer cells. The constitutive activation of erbB-2 and erbB-3 in these cells, coupled with their low, membrane-associated, PTPase activity are likely to play direct roles in driving proliferation of these breast cancer cells.

Prenatal toxicity studies in rats and rabbits with the calcium channel blocker diproteverine.

In prenatal toxicity studies, diproteverine, a calcium channel blocker with demonstrated antianginal properties, produced an unusual pattern of digital, heart, tail, and vertebral defects in rat fetuses from mothers treated during the major period of organogenesis, but only a very low incidence of heart abnormalities was seen in the rabbit. Heart changes were rarely seen in association with digital defects. The findings were consistent with those seen with other calcium channel blockers and add weight to the suggestion of Danielsson and colleagues (5) that digital malformations are a class effect for this type of compound, the effects being related to reduced uteroplacental blood flow. In addition, it is proposed that cardiovascular malformations are also a class response with calcium channel blockers. The distribution of fetal death and hemorrhages and the varying association between cardiovascular, digital, and tail abnormalities seen in the rat with increasing doses of diproteverine fits the pattern of changes reported following hypoxia in the chick embryo. Reduced uteroplacental blood flow with resultant embryonic hypoxia secondary to pharmacologic action is considered a probable mechanism of action for the induction of abnormalities produced by calcium channel blockers.

Computer prediction of possible toxic action from chemical structure: an update on the DEREK system.

Computer-based assessment of potential toxicity has become increasingly popular in recent years. The knowledge-base system DEREK is developed under the guidance of a multinational Collaborative Group of expert toxicologists and provides a qualitative approach to toxicity prediction. Major developments of the DEREK program and knowledge-base have taken place in the last 3 years. Program developments include improvements in both the user interface and data processing. Work on the knowledge-base has concentrated on the areas of genotoxicity and skin sensitisation. DEREK's predictive capabilities for these toxicological end-points has been demonstrated. In addition to the continued expansion of the knowledge-base, a number of enhancements are planned in the DEREK program. In particular, work is in progress to develop further DEREK's ability to report the reasoning behind its predictions.

Reduced expression of the interleukin-2-receptor gamma chain on cord blood lymphocytes: relationship to functional immaturity of the neonatal immune response.

Mutation of the interleukin-2 (IL-2) receptor gamma chain, which also serves as a component of the receptor complexes for IL-4, 7, 9 and 15, results in severe immune deficiency. We hypothesized that the immunological immaturity of healthy neonates might be associated with low levels of expression of this receptor molecule. Using monoclonal antibody and a highly sensitive immunofluorescence method, we showed that IL-2 receptor gamma chain is expressed at significantly lower levels on cord blood cells compared with adult cells. IL-2-dependent T-cell activation in vitro was reduced in cord blood cells compared with adult cells, but B-cell responses to IL-4 were not obviously impaired. The lower level of expression of the gamma chain and some other cytokine receptor chains may contribute to the immunological immaturity of the newborn, by selectively depressing particular immunological mechanisms.

Reorganization of central terminals of myelinated primary afferents in the rat dorsal horn following peripheral axotomy.

We have investigated the time course and extent to which peripheral nerve lesions cause a morphological reorganization of the central terminals of choleragenoid-horseradish peroxidase (B-HRP)-labelled primary afferent fibers in the mammalian dorsal horn. Choleragenoid-horseradish peroxidase is retrogradely transported by myelinated (A) sensory axons to laminae I, III, IV and V of the normal dorsal horn of the spinal cord, leaving lamina II unlabelled. We previously showed that peripheral axotomy results in the sprouting of numerous B-HRP-labelled large myelinated sensory axons into lamina II. We show here that this spread of B-HRP-labelled axons into lamina II is detectable at 1 week, maximal by 2 weeks and persists for over 6 months postlesion. By 9 months, however, B-HRP fibers no longer appear in lamina II. The sprouting into lamina II occurs whether regeneration is allowed (crush) or prevented (section with ligation), and does not reverse at times when peripheral fibers reinnervate the periphery. We also show that 15 times more synaptic terminals in lamina II are labelled by B-HRP 2 weeks after axotomy than in the normal. We interpret this as indicating that the sprouting fibers are making synaptic contacts with postsynaptic targets. This implies that A-fiber terminal reorganization is a prominent and long-lasting but not permanent feature of peripheral axotomy. We also provide evidence that this sprouting is the consequence of a combination of an atrophic loss of central synaptic terminals and the conditioning of the sensory neurons by peripheral axotomy. The sprouting of large sensory fibers into the spinal territory where postsynaptic targets usually receive only small afferent fiber input may bear on the intractable touch-evoked pain that can follow nerve injury.

Cytokine receptor expression in human lymphoid tissue: analysis by fluorescence microscopy.

A highly-sensitive flourescence method, capable of detecting cytokine receptors present at low concentrations (around 100 molecules per cell) by flow cytometry, was adapted for use on tissue sections. This method was used to examine the expression of several cytokine receptors in lymphoid tissues. IL-2 receptors were distributed broadly, with higher concentrations in T cell areas. IL-1 receptor Type 1 was detected in T cell areas and in the follicular mantle, and was strongly expressed on vascular endothelium. IL-6 receptor was found at very low concentration, both within and outside germinal centres. The gp 130 molecule, which is involved in the functional receptor complex for IL-6 and several other cytokines, was present at higher concentrations, particularly in the germinal centre. Analysis of receptor expression in secondary lymphoid tissue provides evidence bearing on the physiological roles of cytokines, as these tissues contain cells at various stages of physiological activation located in well-defined functional zones.

Cytokine receptor expression by solid tumours.

The expression of cytokine receptors by a variety of solid tumour tissues was examined, using an immunofluorescence procedure optimized for sensitivity. Several cytokines generally considered as relevant only to the immune and haematopoietic systems were shown to be expressed by solid tumours. For example, breast carcinoma frequently expressed both chains of the IL-3 receptor, the beta chain of the IL-2 receptor, the TNF type two receptor, the signal-transducing chain CD130, and c-kit. The broad expression of cytokine receptors suggests that the receptor profile of individual tumours should be determined before the application of therapy that involves the administration of cytokines.

Multiple neurotrophic factors including NGF-like activity in nerve regeneration chamber fluids.

Silicone nerve regeneration chambers were implanted between the cut ends of the sciatic nerve of adult rats. Neurotrophic activities in cell-free fluids collected from the chambers were determined using bioassays for survival of embryonic chick ciliary and sympathetic neurons in culture. Separation by molecular exclusion HPLC of the components of fluids collected 1, 2 or 3 days after implantation revealed the presence of a multitude of neurotrophic factors differing in their molecular weights, specificity towards the two types of neurons, and time course. Antiserum to nerve growth factor partially blocked sympathetic activity of fluids collected at 1 day. Affinity purified antibody was also effective and completely eliminated bioactivity of HPLC fractions corresponding to the molecular weight of nerve growth factor. The presence in the fluids of 13-18 and 20-32 kD components active towards ciliary neurons is consistent with the release of fibroblast growth factor and ciliary neurotrophic factor respectively. The stimulation of sympathetic neurons by the 13-18 kD material, and also by 4-6 and 7-11 kD components cannot be entirely accounted for by known factors. This study demonstrates that a number of neurotrophic factors, which differ in their specificity towards sympathetic and parasympathetic neurons, are made available to the region of axonal regrowth over the first few days of regeneration. Contrary to earlier reports, nerve growth factor-like activity was shown to be present in nerve regeneration chambers.

A qualitative assessment of developmental toxicity within a series of structurally related dopamine mimetics.

A qualitative assessment of developmental toxicity within a series of 12 structurally related compounds, 11 of which were active dopamine mimetics and one was inactive, was conducted in rats treated orally by gavage during the major period of organogenesis. Doses were chosen where possible to be equipotent in terms of pharmacological activity. The series was typified by the compound BRL 16644 (2-[[3,4-dihydro-2,2-dimethyl-4-[3-(trifluoromethyl)phenyl]- 2H-1-benzopyran-7-yl]oxy]-N,N-dimethyl-ethanamine: Chemical Abstracts No. 59257-24-8). Five of these compounds were clearly teratogenic producing specific abnormalities typified by anasarca, brachygnathia and cleft palate. Similar levels of maternal toxicity, particularly stereotypic behaviour, and foetotoxicity were seen in both teratogenic and non-teratogenic compounds suggesting that neither maternal nor foetotoxicity plays a role in the aetiology of the abnormalities. Four of the teratogenic compounds contained a trifluoromethyl group in the 4-phenyl ring and, within this series of compounds, substitution with this group appears to confer teratogenicity. Although equipotent doses were used this only pertained to the adult and as only limited pharmacokinetic data were available, including the extent of placental transfer, the influence of this group is not clear. Investigations have been undertaken to relate the teratogenic potential of these compounds to a number of their chemical descriptors, including electronic, steric, quantum chemical and hydrophobicity parameters, to try and clarify the influence of the trifluoromethyl group.

Multivariate quantitative structure-toxicity relationships in a series of dopamine mimetics.

The techniques of principal components analysis and non-linear mapping are routinely used by computer chemists at SmithKline Beecham Pharmaceuticals in the process of drug development by relating the structure of a compound to its chemical activity. To our knowledge these techniques had not previously been applied to the association between the structure of a compound and its toxicological properties. Using a series of 12 structurally related compounds (11 were active dopamine mimetics and one was inactive), of which five were known to be teratogenic and seven were non-teratogenic, it was possible to demonstrate that molecular modelling techniques could be applied to differentiate toxicological data. The structure/property relationships of these compounds were investigated using calculated physicochemical properties, molecular modelling and multivariate statistical techniques. A data set of 56 molecular descriptors was used to represent this series of compounds. Analysis of the data set using principal components analysis and non-linear mapping suggested that teratogenicity was associated with four molecular properties. Moreover, the electronic nature of the 4-phenyl group appeared to be an important determinant of the teratogenesis.