A review of new agents evaluated against pediatric acute lymphoblastic leukemia by the Pediatric Preclinical Testing Program

Acute lymphoblastic leukemia (ALL) in children exemplifies how multi-agent chemotherapy has improved the outcome for patients. Refinements in treatment protocols and improvements in supportive care for this most common pediatric malignancy have led to a cure rate that now approaches 90%. However, certain pediatric ALL subgroups remain relatively intractable to treatment and many patients who relapse face a similarly dismal outcome. Moreover, survivors of pediatric ALL suffer the long-term sequelae of their intensive treatment throughout their lives. Therefore, the development of drugs to treat relapsed/refractory pediatric ALL, as well as those that more specifically target leukemia cells, remains a high priority. As pediatric malignancies represent a minority of the overall cancer burden, it is not surprising that they are generally underrepresented in drug development efforts. The identification of novel therapies relies largely on the reappropriation of drugs developed for adult malignancies. However, despite the large number of experimental agents available, clinical evaluation of novel drugs for pediatric ALL is hindered by limited patient numbers and the availability of effective established drugs. The Pediatric Preclinical Testing Program (PPTP) was established in 2005 to provide a mechanism by which novel therapeutics could be evaluated against xenograft and cell line models of the most common childhood malignancies, including ALL, to prioritize those with the greatest activity for clinical evaluation. In this article, we review the results of >50 novel agents and combinations tested against the PPTP ALL xenografts, highlighting comparisons between PPTP results and clinical data where possible.

Pediatric Preclinical Testing Program (PPTP) testing of the CENP-E inhibitor GSK923295A

10015

Background: GSK923295A is a small molecule inhibitor of centromere-associated protein E (CENP-E), a mitotic kinesin that is required for metaphase chromosome alignment and integration of mitotic spindle mechanics with mitotic checkpoint signaling. An initial phase I clinical trial in adults is ongoing. Methods: The PPTP includes a molecularly characterized in vitro panel of cell lines (n = 27) and in vivo panel of xenografts (n = 60) representing most of the common types of childhood solid tumors and childhood acute lymphoblastic leukemia (ALL). GSK923295A was tested in vitro at concentrations from 1.0 nM to 10.0 μM (96 hour exposure) and was tested in vivo using a daily × 3 for 2 weeks schedule, repeated at day 21. GSK923295A was administered IP at a dose of 125 mg/kg. Three measures of antitumor activity were used: 1) an objective response measure modeled after the clinical setting; 2) a time to event measure based on the median event-free survival (EFS); and 3) a treated to control (T/C) tumor volume measure. Results: GSK923295A demonstrated potent in vitro activity against the PPTP cell line panel with a median IC50 of 27 nM (range 12 nM to > 10 μM). 35 of 37 solid tumor xenograft models were evaluable. GSK923295A induced significant differences in EFS distribution compared to controls in 32 of 35 evaluable models. Objective responses were noted in 13 of 35 xenografts, including 9 with maintained complete responses (MCR), 3 with complete response (CR), and 1 with partial response (PR). Three of 5 Ewing sarcoma xenografts achieved MCR or CR, as did 2 of 3 rhabdoid tumor, and 2 of 5 rhabdomyosarcoma models. For the neuroblastoma panel, the best response was progressive disease (PD) with growth delay compared to controls (PD2 response), which was observed in 5 of 6 xenografts. GSK923295A showed activity against the ALL panel, but unexplained toxicity (generally on or after day 21) precluded formal analysis. Conclusions: GSK923295A has substantial in vitro and in vivo activity against the PPTP's preclinical models. The observed high level of preclinical activity for GSK923295A will need to be evaluated in the context of systemic exposures achieved in the xenograft models and those achievable in humans at tolerable doses. (Supported by NCI NO1CM42216)

[Table: see text]

Pediatric preclinical testing program (PPTP) efficacy and pharmacodynamic evaluation of the Hsp90 inhibitor 17-DMAG

3575

Background: 17-DMAG is a small-molecule inhibitor of the protein chaperone HSP90 that is being developed as an anticancer agent because of the multiple HSP90 client proteins involved in cancer cell growth and survival. Methods: The PPTP includes an in vitro panel (n=27) as well as panels of xenografts (n=61) representing most of the common types of childhood solid tumors and childhood ALL. 17-DMAG was tested against the in vitro panel at concentrations from 1 nM to 10 microM and was tested against the in vivo tumor panels by IP administration using a 50 mg/kg BID twice weekly x 6 weeks dose and schedule. The PPTP’s 3 measures of antitumor activity were used (Houghton et al. Ped Blood Cancer 2006): 1) an objective response measure; 2) treated to control (T/C) tumor volume at day 21; and 3) a time to event (EFS T/C) measure. HSP70 induction was was used as a pharmacodynamic measure of HSP90 inhibition and was determined in tumor and liver tissue at 8 and 24 hours following the second of two doses of 17-DMAG (50 mg/kg IP) administered at 12 hour intervals. Results: 17-DMAG had an EC50 of 62 nM against the PPTP’s in vitro panel, with a trend for lower EC50 values for the rhabdomyosarcoma panel (median EC50 31 nM) compared to the remaining PPTP in vitro cell lines (p=0.06) and for higher EC50 values for the neuroblastoma lines (median EC50 396 nM, p=0.01). 17-DMAG induced significant differences in EFS distribution in 15 of 30 of the solid tumor xenografts, and in 4 of 6 of the evaluable ALL xenografts. Using the time to event activity measure, 17-DMAG had intermediate or high activity against 4 of 28 evaluable solid tumor xenografts (1 of 2 rhabdoid tumor and 3 of 4 alveolar rhabdomyosarcoma). The only objective response (a PR) observed was for an alveolar rhabdomyosarcoma xenograft. HSP70 induction was observed in both liver and tumor tissue, with robust induction (up to 450% increase versus control) occurring in both responding and non-responding tumors. Conclusions: 17-DMAG produced its greatest antitumor activity against alveolar rhabdomyosarcoma xenografts. Robust HSP70 induction was observed in both responding and non-responding xenografts, suggesting that tumor-specific downstream effects of HSP90 inhibition are primary determinants of response. (Supported by NCI NO1CM42216)

No significant financial relationships to disclose.

Local and systemic immune responses to Echinococcus granulosus in experimentally infected dogs

Local and systemic immune responses were studied in six dogs experimentally infected with the dog/sheep tapeworm Echinococcus granulosus. All dogs developed similar IgG antibody response to parasite antigens. In contrast, IgE and IgA responses differed widely. No relationship between IgA responses and parasite burden at the end of the infection were observed. Further, clear differences in the anti-parasite IgA response in serum as compared with specific IgA forming cells in mesenteric lymph nodes were observed within the same dog. An inverse association of anti-parasite IgE and parasite load seemed to be present, with the strongest IgE response in the one dog that had no worms in the intestine at the end of the experiment. No differences were observed in the numbers of intestinal mast cells and goblet cells among all infected dogs. However, the dog with no detectable parasite load had a marked reduction of detected mast cells in the submuscular and muscular layer of the mucosa. Our data give new insight into the immune response of dogs during E. granulosus infection and provide information that may be useful for the rational design of vaccines for the control of hydatid disease.

Avidity analysis of the human immune response to a chitin binding protein of Toxoplasma gondii

A 30-33 kDa electroeluted fraction of T. gondii tachyzoites improved discrimination between acute and chronic phase sera when used instead of the whole tachyzoite extract in an avidity-ELISA. In order to identify the components of these fractions, crude tachyzoite antigen was fractionated by anionic exchange chromatography. The 30-33 kDa antigen cluster eluted in the not-bound fraction could account for a large proportion of the antibody response against the 30-33 kDa electroeluted fraction. According to the N-terminal sequence data, this antigen fraction is composed mainly of SAG1 and another protein with high homology to chitin binding proteins from plants.

Identification of circulating antigens, including an immunoglobulin binding protein, from Toxoplasma gondii tissue cyst and tachyzoites in murine toxoplasmosis

Here we describe the identification of Toxoplasma gondii circulating antigens in sera of BALB/c mice experimentally infected with either the virulent RH strain, or the cystogenic WTD1 strain or with an isolate from a human patient. The circulating antigens were identified by immunoblot in tachyzoite (RH strain) and in tissue cyst (ME-49 strain) crude antigens, using antibodies produced by immunisation of BALB/c mice with homologous sera from infected animals. The most relevant tachyzoite antigen identified are in the following four clusters of 109-94, 67-57, 35-31 and 28-21 kDa. Tissue cyst-specific circulating antigens, like the 18 kDa one, were detected in sera from mice infected with the cystogenic strains. These immune sera, after depletion of tachyzoite specific antibodies, recognised three tissue cysts antigens with Mr of 120, 79 and 48 kDa, and a cluster of antigens in the range of 68-53 kDa. We produced monoclonal antibodies by fusion of myeloma cells with lymphocytes from the mouse immunised with circulating antigens from the RH strain. One of the clones (3A11/H12) obtained, secretes IgG(1) and recognises a peptide epitope from a tachyzoite 67 kDa protein. This parasite protein also binds irrelevant mouse IgG(1) as well as immunoglobulins from other species. The reactivity with non-specific antibodies was inhibited by preincubation with 2% normal mouse and goat serum, while the reaction with the monoclonal antibody 3A11/H12 was not. Furthermore, a biotinylated F(ab')(2) of an irrelevant mouse IgG(1) did not show any reactivity while the F(ab')(2) of the monoclonal antibody 3A11/H12 reacts specifically with the 67 kDa antigen suggesting that this circulating antigen is a putative Fc binding protein.

Salmonella typhimurium as a basis for a live oral Echinococcus granulosus vaccine

A live attenuated Salmonella typhimurium vaccine candidate, LVR01, was constructed by introducing a null deletion into the aroC gene of the parental canine S. typhimurium isolate, P228067. LVR01 was used to orally deliver to the canine immune system a fatty acid binding protein (FABP) from Echinococcus granulosus (EgDf1), as a fusion protein with fragment C (TetC) of tetanus toxin. Immunization studies demonstrated that live LVR01 is well tolerated by orally vaccinated dogs. There was no detectable shedding of the vaccine strain in the faeces 2 days after immunization. Humoral antibody responses were observed against Salmonella, TetC and EgDf1. Cellular responses were consistently detected against Salmonella and TetC. A cellular response against EgDf1 was also seen in a proportion of the LVR01 vaccinated dogs. We propose S. typhimurium LVR01 as a carrier for recombinant antigens and a vector for the construction of multivalent oral vaccines for dogs.

A mucosal IgA response, but no systemic antibody response, is evoked by intranasal immunisation of dogs with Echinococcus granulosus surface antigens iscoms

The search for protective antigens of intestinal parasites is conditioned by the methodology used to induce a relevant local immune response against them. The present work describes the use of immuno stimulating complexes (iscoms) from tegumental antigens from protoscoleces (PSC) of the cestode Echinococcus granulosus as immunogens in dogs by the intranasal route. It also describes the evaluation of the immune response evoked at the antibody level (systemically and at a distant mucosal location) as well as at the level of antibody secreting cells in peripheral blood. Iscoms from both E. granulosus tegumental antigens and hen ovalbumin (OVA), given at 50 microg doses by intranasal route, evoked significant secretory IgA antibody responses detected in saliva. Specific IgA secreting cells in peripheral blood also increased 10-20-fold, although transiently, after primary and secondary stimulation, whereas specific IgG secreting cells in peripheral blood were only detected in some individuals after the second antigenic exposure. Generation of immune responses at a related mucosal site provides evidence of localised immunity. No significant increase in systemic antibody titers of either IgM, IgG or IgA isotype was detected in plasma as a result of the immunisation. This fact could reflect that the nasopharyngeal mucosal associated lymphoid tissue of dogs is more strictly compartmentalised than that of other mammals.

Fc-binding molecules specific for human IgG1 and IgG3 are present in Echinococcus granulosus protoscoleces

In this work we describe the presence of Fc-binding activity on the suckers and tegument of E. granulosus protoscoleces. A fraction (PSA-Fc+) from protoscolex somatic antigens was isolated by affinity chromatography on human Fc-gamma1-Sepharose. Analysis by SDS-PAGE of PSA-Fc+ showed that it contained two major components. Using mouse F(ab')2-human Fc chimaeric monoclonal antibodies we verified that PSA-Fc+ bound mainly to human Fc-gamma1 and Fc-gamma3 isotypes. In addition, one of the components of PSA-Fc+ showed proteolytic activity. Both, Fc-binding and proteolytic activities localized on the protoscolex surface, may play a relevant role in the host-parasite interaction.

Evolution of IgG antibody response against Toxoplasma gondii tissue cyst in acute and chronic human infections

The recognition profile of the tissue cysts antigens by IgG antibodies was studied during acute and chronic human toxoplasmic infection. Thus the IgG response against Toxoplasma gondii was investigated by immunoblotting in two patients accidentally infected with the RH strain as well as in group of naturally infected patients at acute and chronic phase. There was an overall coincidence of molecular mass among antigens of tachyzoites and tissue cysts recognized by these sera, however, they appear not to be the same molecules. The response against tissue cysts starts early during acute infection, and the reactivity of antibodies is strong against a wide range of antigens. Six bands (between 82 and 151 kDa) were exclusively recognized by chronic phase sera but only the 132 kDa band was positive in more than 50% of the sera analysed. A mixture of these antigens could be used to discriminate between the two infection phases. The most important antigens recognized by the acute and the chronic phase sera were 4 clusters in the ranges 20-24 kDa, 34-39 kDa, 58-80 kDa and 105-130 kDa as well as two additional antigens of 18 and 29 kDa. Both accidentally infected patients and some of the naturally infected patients showed a weak specific response against tissue cyst antigens.

"PERFEXT": a direct method for quantitative assessment of cytokine production in vivo at the local level

A method termed "PERFEXT", based on sequential perfusion and detergent extraction of lymphoid and non-lymphoid organs, has been developed for the quantitative measurement of cytokines produced at a local level in a given tissue. In vivo treatment of mice with Staphylococcus enterotoxin B (SEB) or lipopolysaccharide (LPS) served as the model systems. Interleukin-2 (IL2) and interferon-gamma (IFN gamma) levels were monitored by ELISA analysis of extracted samples. After local footpad (FP) injection with SEB, spleen and serum IL2 levels peaked at 2-4 h, while IL2 levels peaked at around 4-8 h in both FP and popliteal lymph nodes. SEB injection resulted in increased IFN gamma levels both in the FP and the draining lymph node. The detection of cytokines in the intestine allows for the application of the method at mucosal sites as well, provided enzyme inhibitors are present during the extraction procedure. After FP injection with LPS, IFN gamma production was significantly increased in the draining lymph node and was detectable in the FP, whereas IL2 was undetectable in any organ examined. IL2 and IFN gamma could also be detected at the site of elicitation of a delayed-type hypersensitivity reaction following local FP challenge. Local cytokine production correlated with the swelling response, whereas cytokine production in the spleen did not. IL2 peaked early, followed by a late increase in IFN gamma production, corresponding to the maximum swelling. This simple method should prove useful for analysing the production of other cytokines in vivo in distinct anatomical compartments.

Intranasal immunization of mice with Echinococcus granulosus surface antigens iscoms evokes a strong immune response, biased towards glucidic epitopes

The present work describes the preparation and characterization of iscoms from tegumental antigens of Echinococcus granulosus protoscoleces, and their use as immunogens in mice by intranasal and subcutaneous routes. Iscoms given at 10 mg doses evoke a significant antibody response when administered i.n. and a strong and long lasting antibody response by s.c. route. The Ag administered i.n. in a non-adjuvanted form as a booster was not immunogenic. The i.n. route of immunization induced higher IgA serum titre in relation to IgG than the s.c. route and antisera of slightly higher avidity. Also, ten fold lower IgG2a/IgG1 and ten fold higher IgG3/IgG1 ratios were observed for the i.n. protocol compared to the s.c. one. Moreover, the mucosal route of immunization induced more efficiently antibodies of both isotypes directed to carbohydrate epitopes. The differences between immune response generated by i.n. and parenteral immunizations should be taken into consideration, particularly in those cases in which protective antigens are glycosylated.

Idiotypic modulation of the antibody response of mice to Echinococcus granulosus antigens

In this work we analysed the modulation of the antibody response to Echinococcus granulosus antigens via anti-idiotype (Ab2) administration. In a first set of experiments, we determined the antibody response to hydatid cyst fluid antigen (HCFA) of mice immunized with Ab2. Results showed that Ab2 elicited anti-HCFA antibodies in mice that had never been exposed to HCFA before. Moreover, that response was characterized by booster effect, avidity maturation and an inverse correlation between avidity and Ab2 dose. We infected these mice and measured the titre and avidity of anti-HCFA antibodies during 250 days of infection. Lack of avidity maturation was the most important feature observed, suggesting that the parasite (either protoscolex or cyst) could modulate the antibody response of its host. In a second set of experiments, we investigated the presence of regulatory anti-idiotypes in the Ab2 preparation. In this context we treated neonates with different doses of Ab2, immunized them with HCFA 12 weeks later and determined their anti-HCFA titres. A specific and Ab2 dose-dependent suppression of the response to HCFA was observed, suggesting the existence of regulatory anti-idiotypes in the Ab2 preparation. Although this Ab2 could be involved in the regulation of the anti-HCFA response, other anti-idiotypes could also be involved. Our results show that it may be possible to improve the avidity of the anti-HCFA response via the administration of the anti-idiotype Ab. However, the live parasite could successfully revert this effect by mechanisms not yet characterized.

Specific glycosylation site mutations of the insulin receptor alpha subunit impair intracellular transport

The insulin receptor is a transmembrane protein found on multiple cell types. This receptor is synthesized as a 190-kDa proreceptor which is cleaved to produce mature alpha and beta subunits. The proreceptor contains 18 potential sites for N-linked glycosylation: 14 on the alpha subunit and 4 on the beta subunit. The codons for asparagine in the first four sites at the amino terminus of the alpha subunit were mutated to code for glutamine. This mutant receptor cDNA was stably transfected into NIH 3T3 cells. The insulin receptor produced in these cells remained in the proreceptor form; no mature alpha and beta subunits were produced. The proreceptor was slightly smaller on SDS-PAGE gels than the wild-type proreceptor and contained four less oligosaccharide chains by tryptic peptide mapping. The carbohydrate chains on the mutant proreceptor remained endoglycosidase H sensitive. However, in the presence of brefeldin A, these oligosaccharide chains could be processed to endoglycosidase H resistant chains. By immunofluorescence, the mutant proreceptor was shown to be localized to the endoplasmic reticulum. No insulin receptors could be found on the cell-surface either with cell surface labeling with biotin or with 125I-insulin binding. Thus, glycosylation of the first four N-linked glycosylation sites of the insulin receptor is necessary for the proper processing and intracellular transport of the receptor. This is in contrast to glycosylation at the four sites on the beta subunit which appear not to be important for processing but necessary for signal transduction. Therefore, N-linked glycosylation of the insulin receptor at specific sites has multiple distinctive roles.

Lack of interspecies barriers in anti-Id stimulated antibody production against Echinococcus granulosus antigens

Polyclonal human anti-hydatid antibodies were affinity purified from a hydatid patient serum and used to produce a rabbit anti-idiotypic serum. These anti-Id antibodies cross-reacted in ELISA with sera from 11 of 12 hydatid patients studied and with 13 infected or immunized mice sera. All mice primed and boosted with anti-Id produced anti-hydatid antibodies in the primary response and exhibited an increase in antibody titre after a booster injection. The same effect was observed with mice primed with antigen and boosted with anti-Id, although these mice exhibited higher antibody titres. A significant idiotype repertoire is shared by anti-hydatid antibodies produced by different individuals of the same or different species, and anti-Id raised against those antibodies behave as surrogate antigens producing a normal primary and secondary response in animals of different species from that used to isolate the Id.