Statistical Approach to Identify Threshold and Point of Departure in Dose-Response Data.

The study presents an integrated, rigorous statistical approach to define the likelihood of a threshold and point of departure (POD) based on dose-response data using nested family of bent-hyperbola models. The family includes four models: the full bent-hyperbola model, which allows for transition between two linear regiments with various levels of smoothness; a bent-hyperbola model reduced to a spline model, where the transition is fixed to a knot; a bent-hyperbola model with a restricted negative asymptote slope of zero, named hockey-stick with arc (HS-Arc); and spline model reduced further to a hockey-stick type model (HS), where the first linear segment has a slope of zero. A likelihood-ratio test is used to discriminate between the models and determine if the more flexible versions of the model provide better or significantly better fit than a hockey-stick type model. The full bent-hyperbola model can accommodate both threshold and nonthreshold behavior, can take on concave up and concave down shapes with various levels of curvature, can approximate the biochemically relevant Michaelis-Menten model, and even be reduced to a straight line. Therefore, with the use of this model, the presence or absence of a threshold may even become irrelevant and the best fit of the full bent-hyperbola model be used to characterize the dose-response behavior and risk levels, with no need for mode of action (MOA) information. Point of departure (POD), characterized by exposure level at which some predetermined response is reached, can be defined using the full model or one of the better fitting reduced models.

A Phase I Study of Unimolecular Pentavalent (Globo-H-GM2-sTn-TF-Tn) Immunization of Patients with Epithelial Ovarian, Fallopian Tube, or Peritoneal Cancer in First Remission.

We conducted a phase I study in ovarian cancer patients to evaluate the safety and immunogenicity of a synthetic unimolecular pentavalent carbohydrate vaccine (Globo-H, GM2, sTn, TF, and Tn) supported on a peptide backbone, conjugated to keyhole limpet haemocyanin (KLH), and mixed with immunological adjuvant QS-21. Twenty-four advanced-stage, poor-risk, first-remission ovarian cancer patients were enrolled from January 2011-Septermber 2013. Three dose levels were planned (25, 50, 100 mcg) with three cohorts of six patients each, with an additional 6-patient expansion cohort at the MTD. ELISA serologic IgM and IgG responses for each antigen was defined as positive response if antibody titers were ≥1:80 over the respective patient's pre-vaccination serum. The study would be considered positive if at least four of 12 patients treated at the MTD showed immune responses for at least three of the five antigens. Twenty-four patients (median age, 54 years [range, 36-68]) were included in the safety analysis. Histology was high-grade serous in 22 patients (92%); 18 had stage III and six stage IV disease. The vaccine was well-tolerated at all doses, with no DLTs. At the highest treated dose, IgG and/or IgM responses were recorded against ≥3 antigens in 9/12 patients (75%), ≥4 in 7/12 (58%), and 5 in 3/12 (25%). With a median follow-up of 19 months (range, 2-39), 20 patients (83%) recurred and six (25%) died. The unimolecular pentavalent vaccine construct was shown to be safe and immunogenic. Such a construct greatly simplifies regulatory requirements and manufacturing, facilitates scalability, and provides adaptability.

Kinetic Modeling Reveals the Roles of Reactive Oxygen Species Scavenging and DNA Repair Processes in Shaping the Dose-Response Curve of KBrO3-Induced DNA Damage.

We have developed a kinetic model to investigate how DNA repair processes and scavengers of reactive oxygen species (ROS) can affect the dose-response shape of prooxidant induced DNA damage. We used as an example chemical KBrO3 which is activated by glutathione and forms reactive intermediates that directly interact with DNA to form 8-hydroxy-2-deoxyguanosine DNA adducts (8-OH-dG). The single strand breaks (SSB) that can result from failed base excision repair of these adducts were considered as an effect downstream from 8-OH-dG. We previously demonstrated that, in the presence of effective base excision repair, 8-OH-dG can exhibit threshold-like dose-response dependence, while the downstream SSB can still exhibit a linear dose-response. Here we demonstrate that this result holds for a variety of conditions, including low levels of GSH, the presence of additional SSB repair mechanisms, or a scavenger. It has been shown that melatonin, a terminal scavenger, inhibits KBrO3-caused oxidative damage. Our modeling revealed that sustained exposure to KBrO3 can lead to fast scavenger exhaustion, in which case the dose-response shapes for both endpoints are not substantially affected. The results are important to consider when forming conclusions on a chemical's toxicity dose dependence based on the dose-response of early genotoxic events.

Systematic proteomic approach to characterize the impacts of chemical interactions on protein and cytotoxicity responses to metal mixture exposures.

Chemical interactions have posed a big challenge in toxicity characterization and human health risk assessment of environmental mixtures. To characterize the impacts of chemical interactions on protein and cytotoxicity responses to environmental mixtures, we established a systems biology approach integrating proteomics, bioinformatics, statistics, and computational toxicology to measure expression or phosphorylation levels of 21 critical toxicity pathway regulators and 445 downstream proteins in human BEAS-2B cells treated with 4 concentrations of nickel, 2 concentrations each of cadmium and chromium, as well as 12 defined binary and 8 defined ternary mixtures of these metals in vitro. Multivariate statistical analysis and mathematical modeling of the metal-mediated proteomic response patterns showed a high correlation between changes in protein expression or phosphorylation and cellular toxic responses to both individual metals and metal mixtures. Of the identified correlated proteins, only a small set of proteins including HIF-1α is likely to be responsible for selective cytotoxic responses to different metals and metals mixtures. Furthermore, support vector machine learning was utilized to computationally predict protein responses to uncharacterized metal mixtures using experimentally generated protein response profiles corresponding to known metal mixtures. This study provides a novel proteomic approach for characterization and prediction of toxicities of metal and other chemical mixtures.

Nonparametric Bayesian methods for benchmark dose estimation.

The article proposes and investigates the performance of two Bayesian nonparametric estimation procedures in the context of benchmark dose estimation in toxicological animal experiments. The methodology is illustrated using several existing animal dose-response data sets and is compared with traditional parametric methods available in standard benchmark dose estimation software (BMDS), as well as with a published model-averaging approach and a frequentist nonparametric approach. These comparisons together with simulation studies suggest that the nonparametric methods provide a lot of flexibility in terms of model fit and can be a very useful tool in benchmark dose estimation studies, especially when standard parametric models fail to fit to the data adequately.

Dose-response analysis of bromate-induced DNA damage and mutagenicity is consistent with low-dose linear, nonthreshold processes.

Mutagenic agents have long been inferred to act through low-dose linear, nonthreshold processes. However, there is debate about this assumption, with various studies interpreting datasets as showing thresholds for DNA damage and mutation. We have applied rigorous statistical analyses to investigate the shape of dose-response relationships for a series of in vitro and in vivo genotoxicity studies using potassium bromate (KBrO(3) ), a water ozonation byproduct that is bioactivated to a reactive species causing oxidative damage to DNA. We analyzed studies of KBrO(3) genotoxicity where no-effect/threshold levels were reported as well as other representative datasets. In all cases, the data were consistent with low-dose linear models. In the majority of cases, the data were fit either by a linear (straight line) model or a model which was linear at low doses and showed a saturation-like downward curvature at high doses. Other datasets with apparent upward curvature were still adequately represented by models that were linear at low dose. Sensitivity analysis of datasets showing upward curvature revealed that both low-dose linear and nonlinear models provide adequate fits. Additionally, a simple biochemical model of selected key processes in bromate-induced DNA damage was developed and illustrated a situation where response for early primary events suggested an apparent threshold while downstream events were linear. Overall, the statistical analyses of DNA damage and mutations induced by KBrO(3) are consistent with a low-dose linear response and do not provide convincing evidence for the presence of a threshold.

Dual targeting of EGFR can overcome a major drug resistance mutation in mouse models of EGFR mutant lung cancer.

EGFR is a major anticancer drug target in human epithelial tumors. One effective class of agents is the tyrosine kinase inhibitors (TKIs), such as gefitinib and erlotinib. These drugs induce dramatic responses in individuals with lung adenocarcinomas characterized by mutations in exons encoding the EGFR tyrosine kinase domain, but disease progression invariably occurs. A major reason for such acquired resistance is the outgrowth of tumor cells with additional TKI-resistant EGFR mutations. Here we used relevant transgenic mouse lung tumor models to evaluate strategies to overcome the most common EGFR TKI resistance mutation, T790M. We treated mice bearing tumors harboring EGFR mutations with a variety of anticancer agents, including a new irreversible EGFR TKI that is under development (BIBW-2992) and the EGFR-specific antibody cetuximab. Surprisingly, we found that only the combination of both agents together induced dramatic shrinkage of erlotinib-resistant tumors harboring the T790M mutation, because together they efficiently depleted both phosphorylated and total EGFR. We suggest that these studies have immediate therapeutic implications for lung cancer patients, as dual targeting with cetuximab and a second-generation EGFR TKI may be an effective strategy to overcome T790M-mediated drug resistance. Moreover, this approach could serve as an important model for targeting other receptor tyrosine kinases activated in human cancers.

From synthesis to biologics: preclinical data on a chemistry derived anticancer vaccine.

A fully synthetic anticancer vaccine 2 has been prepared via bioconjugation of unimolecular pentavalent construct 1-containing five prostate and breast cancer associated carbohydrate antigens, Globo-H, GM2, STn, TF and Tn-to maleimide-modified carrier protein KLH. An improved conjugation protocol has been developed, which allowed us to obtain a higher epitope ratio of the unimolecular pentavalent glycopeptide antigen to the carrier protein (505/1 versus 228/1 for the previous version). KLH conjugate 2 has been subsequently submitted to preclinical immunogenic evaluation in mice in the presence of QS-21 as an adjuvant. Through standard ELISA assay, this vaccine candidate showed high promise in inducing IgG and IgM antibodies against each of the five individual carbohydrate antigens. In addition, FACS analysis indicated that these antibodies were able to react with MCF-7 breast cancer cell lines expressing these five carbohydrate antigens.

Voltage gating at the selectivity filter of the Ca2+ release-activated Ca2+ channel induced by mutation of the Orai1 protein.

The Ca(2+) release-activated Ca(2+) (CRAC) channel is a plasma membrane (PM) channel that is uniquely activated when free Ca(2+) level in the endoplasmic reticulum (ER) is substantially reduced. Several small interfering RNA screens identified two membrane proteins, Orai1 and STIM1, to be essential for the CRAC channel function. STIM1 appears to function in the PM and as the Ca(2+) sensor in the ER. Orai1 is forming the pore of the CRAC channel. Despite the recent breakthroughs, a mechanistic understanding of the CRAC channel gating is still lacking. Here we reveal new insights on the structure-function relationship of STIM1 and Orai1. Our data suggest that the cytoplasmic coiled-coil region of STIM1 provides structural means for coupling of the ER membrane to the PM to activate the CRAC channel. We mutated two hydrophobic residues in this region to proline (L286P/L292P) to introduce a kink in the first alpha-helix of the coiled-coil domain. This STIM1 mutant caused a dramatic inhibition of the CRAC channel gating compared with the wild type. Structure-function analysis of the Orai1 protein revealed the presence of intrinsic voltage gating of the CRAC channel. A mutation of Orai1 (V102I) close to the selectivity filter modified CRAC channel voltage sensitivity. Expression of the Orai1(V102I) mutant resulted in slow voltage gating of the CRAC channel by negative potentials. The results revealed that the alteration of Val(102) develops voltage gating in the CRAC channel. Our data strongly suggest the presence of a novel voltage gating mechanism at the selectivity filter of the CRAC channel.

Pilot study of a heptavalent vaccine-keyhole limpet hemocyanin conjugate plus QS21 in patients with epithelial ovarian, fallopian tube, or peritoneal cancer.

PURPOSE: To characterize the safety and immunogenicity of a heptavalent antigen-keyhole limpet hemocyanin (KLH) plus QS21 vaccine construct in patients with epithelial ovarian, fallopian tube, or peritoneal cancer in second or greater complete clinical remission. EXPERIMENTAL DESIGN: Eleven patients in this pilot trial received a heptavalent vaccine s.c. containing GM2 (10 microg), Globo-H (10 microg), Lewis Y (10 microg), Tn(c) (3 microg), STn(c) (3 microg), TF(c) (3 microg), and Tn-MUC1 (3 microg) individually conjugated to KLH and mixed with adjuvant QS21(100 microg). Vaccinations were administered at weeks 1, 2, 3, 7, and 15. Periodic blood and urine samples were obtained to monitor safety (complete blood count, comprehensive panel, amylase, thyroid-stimulating hormone, and urinalysis) and antibody production (ELISA, fluorescence-activated cell sorting, and complement-dependent cytotoxicity). RESULTS: Eleven patients were included in the safety analysis; 9 of 11 patients remained on study for at least 2 weeks past fourth vaccination and were included in the immunologic analysis (two withdrew, disease progression). The vaccine was well tolerated. Self-limited and mild fatigue (maximum grade 2 in two patients), fever, myalgia, and localized injection site reactions were most frequent. No clinically relevant hematologic abnormalities were noted. No clinical or laboratory evidence of autoimmunity was seen. Serologic responses by ELISA were largely IgM against each antigen with the exception of Tn-MUC1 where both IgM and IgG responses were induced. Antibody responses were generally undetectable before immunization. After immunization, median IgM titers were as follows: Tn-MUC1, 1:640 (IgG 1:80); Tn, 1:160; TF, 1:640; Globo-H, 1:40; and STn, 1:80. Only one response was seen against Lewis Y; two were against GM2. Eight of nine patients developed responses against at least three antigens. Antibody titers peaked at weeks 4 to 8 in all patients. Fluorescence-activated cell sorting and complement-dependent cytotoxicity analysis showed substantially increased reactivity against MCF7 cells in seven of nine patients, with some increase seen in all patients. CONCLUSIONS: This heptavalent-KLH conjugate plus QS21 vaccine safely induced antibody responses against five of seven antigens. Investigation in an adequately powered efficacy trial is warranted.

Calcium signals mediated by STIM and Orai proteins--a new paradigm in inter-organelle communication.

In all cells Ca2+ signals are key to controlling a spectrum of cellular responses. Ca2+ signals activated by phospholipase C-coupled receptors have two components-rapid Ca2+ release from ER stores followed by slower Ca2+ entry from outside the cell. The coupling process between ER and PM to mediate this "store-operated" Ca2+ entry process has remained a molecular and mechanistic mystery. Through a combination of high throughput screening and molecular physiological approaches, the machinery and mechanism of this process have been elucidated. Two proteins are key to the coupling process. STIM1, a single spanning membrane protein with an unpaired Ca2+ binding EF-hand functions as the sensor of ER luminal Ca2+ and through redistribution in the ER transduces information directly to the PM. Orai1, a tetra-spanning PM protein, functions as the highly Ca2+ selective channel in the PM that is gated through interactions with the store-activated ER Ca2+ sensor. This molecular pas-de-deux between ER and PM components represents not only a crucial signaling pathway, but also a new paradigm in inter-organelle communication.

A common mechanism underlies stretch activation and receptor activation of TRPC6 channels.

The TRP family of ion channels transduce an extensive range of chemical and physical signals. TRPC6 is a receptor-activated nonselective cation channel expressed widely in vascular smooth muscle and other cell types. We report here that TRPC6 is also a sensor of mechanically and osmotically induced membrane stretch. Pressure-induced activation of TRPC6 was independent of phospholipase C. The stretch responses were blocked by the tarantula peptide, GsMTx-4, known to specifically inhibit mechanosensitive channels by modifying the external lipid-channel boundary. The GsMTx-4 peptide also blocked the activation of TRPC6 channels by either receptor-induced PLC activation or by direct application of diacylglycerol. The effects of the peptide on both stretch- and diacylglycerol-mediated TRPC6 activation indicate that the mechanical and chemical lipid sensing by the channel has a common molecular mechanism that may involve lateral-lipid tension. The mechanosensing properties of TRPC6 channels highly expressed in smooth muscle cells are likely to play a key role in regulating myogenic tone in vascular tissue.

STIM2 is an inhibitor of STIM1-mediated store-operated Ca2+ Entry.

The coupling mechanism between endoplasmic reticulum (ER) Ca(2+) stores and plasma membrane (PM) store-operated channels (SOCs) remains elusive [1-3]. STIM1 was shown to play a crucial role in this coupling process [4-7]; however, the role of the closely related STIM2 protein remains undetermined. We reveal that STIM2 is a powerful SOC inhibitor when expressed in HEK293, PC12, A7r5, and Jurkat T cells. This contrasts with gain of SOC function in STIM1-expressing cells. While STIM1 is expressed in both the ER and plasma membrane, STIM2 is expressed only intracellularly. Store depletion induces redistribution of STIM1 into distinct "puncta." STIM2 translocates into puncta upon store depletion only when coexpressed with STIM1. Double labeling shows coincidence of STIM1 and STIM2 within puncta, and immunoprecipitation reveals direct interactions between STIM1 and STIM2. Independent of store depletion, STIM2 colocalizes with and blocks the function of a STIM1 EF-hand mutant that preexists in puncta and is constitutively coupled to activate SOCs. Thus, whereas STIM1 is a required mediator of SOC activation, STIM2 is a powerful inhibitor of this process, interfering with STIM1-mediated SOC activation at a point downstream of puncta formation. The opposing functions of STIM1 and STIM2 suggest they may play a coordinated role in controlling SOC-mediated Ca(2+) entry signals.

Orai1 and STIM reconstitute store-operated calcium channel function.

The two membrane proteins, STIM1 and Orai1, have each been shown to be essential for the activation of store-operated channels (SOC). Yet, how these proteins functionally interact is not known. Here, we reveal that STIM1 and Orai1 expressed together reconstitute functional SOCs. Expressed alone, Orai1 strongly reduces store-operated Ca(2+) entry (SOCE) in human embryonic kidney 293 cells and the Ca(2+) release-activated Ca(2+) current (I(CRAC)) in rat basophilic leukemia cells. However, expressed along with the store-sensing STIM1 protein, Orai1 causes a massive increase in SOCE, enhancing the rate of Ca(2+)entry by up to 103-fold. This entry is entirely store-dependent since the same coexpression causes no measurable store-independent Ca(2+) entry. The entry is completely blocked by the SOC blocker, 2-aminoethoxydiphenylborate. Orai1 and STIM1 coexpression also caused a large gain in CRAC channel function in rat basophilic leukemia cells. The close STIM1 homologue, STIM2, inhibited SOCE when expressed alone but coexpressed with Orai1 caused substantial constitutive (store-independent) Ca(2+) entry. STIM proteins are known to mediate Ca(2+) store-sensing and endoplasmic reticulum-plasma membrane coupling with no intrinsic channel properties. Our results revealing a powerful gain in SOC function dependent on the presence of both Orai1 and STIM1 strongly suggest that Orai1 contributes the PM channel component responsible for Ca(2+) entry. The suppression of SOC function by Orai1 overexpression likely reflects a required stoichiometry between STIM1 and Orai1.

STIM1 has a plasma membrane role in the activation of store-operated Ca(2+) channels.

Receptor-induced Ca(2+) signals are key to the function of all cells and involve release of Ca(2+) from endoplasmic reticulum (ER) stores, triggering Ca(2+) entry through plasma membrane (PM) "store-operated channels" (SOCs). The identity of SOCs and their coupling to store depletion remain molecular and mechanistic mysteries. The single transmembrane-spanning Ca(2+)-binding protein, STIM1, is necessary in this coupling process and is proposed to function as an ER Ca(2+) sensor to provide the trigger for SOC activation. Here we reveal that, in addition to being an ER Ca(2+) sensor, STIM1 functions within the PM to control operation of the Ca(2+) entry channel itself. Increased expression levels of STIM1 correlate with a gain in function of Ca(2+) release-activated Ca(2+) (CRAC) channel activity. Point mutation of the N-terminal EF hand transforms the CRAC channel current (I(CRAC)) into a constitutively active, Ca(2+) store-independent mode. Mutants in the EF hand and cytoplasmic C terminus of STIM1 alter operational parameters of CRAC channels, including pharmacological profile and inactivation properties. Last, Ab externally applied to the STIM1 N-terminal EF hand blocks both I(CRAC) in hematopoietic cells and SOC-mediated Ca(2+) entry in HEK293 cells, revealing that STIM1 has an important functional presence within the PM. The results reveal that, in addition to being an ER Ca(2+) sensor, STIM1 functions within the PM to exert control over the operation of SOCs. As a cell surface signaling protein, STIM1 represents a key pharmacological target to control fundamental Ca(2+)-regulated processes including secretion, contraction, metabolism, cell division, and apoptosis.

Preparation and evaluation of unimolecular pentavalent and hexavalent antigenic constructs targeting prostate and breast cancer: a synthetic route to anticancer vaccine candidates.

Several novel, fully synthetic, carbohydrate-based antitumor vaccines have been assembled. Each construct consists of multiple cancer-related antigens displayed on a single polypeptide backbone. Recent advances in synthetic methodology have allowed for the incorporation of a complex oligosaccharide terminating in a sialic acid residue (i.e., GM2) as one of the carbohydrate antigens. Details of the vaccine synthesis as well as the results of preliminary immunological investigations are described herein.

Role of endogenous TRPC6 channels in Ca2+ signal generation in A7r5 smooth muscle cells.

The ubiquitously expressed canonical transient receptor potential (TRPC) ion channels are considered important in Ca2+ signal generation, but their mechanisms of activation and roles remain elusive. Whereas most studies have examined overexpressed TRPC channels, we used molecular, biochemical, and electrophysiological approaches to assess the expression and function of endogenous TRPC channels in A7r5 smooth muscle cells. Real time PCR and Western analyses reveal TRPC6 as the only member of the diacylglycerol-responsive TRPC3/6/7 subfamily of channels expressed at significant levels in A7r5 cells. TRPC1, TRPC4, and TRPC5 were also abundant. An outwardly rectifying, nonselective cation current was activated by phospholipase C-coupled vasopressin receptor activation or by the diacylglycerol analogue, oleoyl-2-acetyl-sn-glycerol (OAG). Introduction of TRPC6 small interfering RNA sequences into A7r5 cells by electroporation led to 90% reduction of TRPC6 transcript and 80% reduction of TRPC6 protein without any detectable compensatory changes in the expression of other TRPC channels. The OAG-activated nonselective cation current was similarly reduced by TRPC6 RNA interference. Intracellular Ca2+ measurements using fura-2 revealed that thapsigargin-induced store-operated Ca2+ entry was unaffected by TRPC6 knockdown, whereas vasopressin-induced Ca2+ entry was suppressed by more than 50%. In contrast, OAG-induced Ca2+ transients were unaffected by TRPC6 knockdown. Nevertheless, OAG-induced Ca2+ entry bore the hallmarks of TRPC6 function; it was inhibited by protein kinase C and blocked by the Src-kinase inhibitor, 4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo[3,4-d]pyrimidine (PP2). Importantly, OAG-induced Ca2+ entry was blocked by the potent L-type Ca2+ channel inhibitor, *nimodipine. Thus, TRPC6 activation probably results primarily in Na ion entry and depolarization, leading to activation of L-type channels as the mediators of Ca2+ entry. Calculations reveal that even 90% reduction of TRPC6 channels would allow depolarization sufficient to activate L-type channels. This tight coupling between TRPC6 and L-type channels is probably important in mediating smooth muscle cell membrane potential and muscle contraction.

Synthesis of selected LeY and KH-1 analogues: a medicinal chemistry approach to vaccine optimization.

[structure: see text] As part of our ongoing anticancer vaccine program, we recently found that antibodies generated in response to the KH-1-KLH construct recognized not only KH-1 antigen but also the Lewis Y (Le(y)) antigen as well, with antibody titer levels much higher than those observed after immunization with individual Le(y)-KLH vaccine constructs. In an attempt to explore the structure-antigenic relationship of these carbohydrate epitopes, several analogues of both KH-1 and Le(y) were synthesized. A convergent synthetic approach to the analogues was designed on the basis of well-established glycal methodology, employing a minimum number of building blocks to generate competent antigens with high stereoselectivity and reasonable yield.

A bivalent conjugate vaccine in the treatment of biochemically relapsed prostate cancer: a study of glycosylated MUC-2-KLH and Globo H-KLH conjugate vaccines given with the new semi-synthetic saponin immunological adjuvant GPI-0100 OR QS-21.

GPI-0100 is a semi-synthetic saponin with modifications designed to augment stability and diminish toxicity. Two batches of GPI-0100 (the second with higher purity) were tested with doses ranging between 100 and 5000 microg in groups of five treated prostate cancer patients who had no evidence of disease except for rising PSA levels. GPI-0100 was mixed with a bivalent vaccine containing the glycolipid Globo H and the glycosylated mucin MUC2 conjugated to keyhole limpet hemocyanin (KLH). All doses were well tolerated and antibody titers against Globo H and MUC-2 escalated with the increasing dose levels. At the 5000 microg dose level in this patient population, toxicity remained minimal with only occasional grade II local toxicity at vaccination sites and occasional sporadic grade I elevations in ALT. Compared with a subsequent trial with the same bivalent vaccine plus QS-21 at the maximal tolerated dose of 100 microg, the 5000 microg dose of GPI-0100 produced comparable antibody titers.