Article Tools
Phase I and Clinical Pharmacology
Article Tools
Phase I Trial of 17-Allylamino-17-Demethoxygeldanamycin in Patients With Advanced Cancer
Abstract
We determined the maximum-tolerated dose (MTD) and the dose-limiting toxicities (DLT) of 17-allylamino-17-demethoxygeldanamycin (17-AAG) when infused on days 1, 8, and 15 of a 28-day cycle in advanced solid tumor patients. We also characterized the pharmacokinetics of 17-AAG, its effect on chaperone and client proteins, and whether cytochrome P450 (CYP) 3A5 and NAD(P)H:quinone oxidoreductase 1 (NQO1) polymorphisms affected 17-AAG disposition or toxicity.
An accelerated titration design was used. Biomarkers were measured in peripheral-blood mononuclear cells (PBMCs) at baseline and on days 1 and 15, and pharmacokinetic analysis was performed on day 1 of cycle 1. CYP3A5*3 and NQO1*2 genotypes were determined and correlated with pharmacokinetics and toxicity.
Twenty-one patients received 52 courses at 11 dose levels. DLTs at 431 mg/m2 were grade 3 bilirubin (n = 1), AST (n = 1), anemia (n = 1), nausea (n = 1), vomiting (n = 1), and myalgias (n = 1). No tumor responses were seen. 17-AAG consistently increased heat shock protein (Hsp) 70 levels in PBMCs. At the MTD, the clearance and half-life (t1/2) of 17-AAG were 11.6 L/h/m2 and 4.15 hours, respectively; whereas the active metabolite 17-aminogeldanamycin had a t1/2 of 7.63 hours. The CYP3A5*3 and NQO1*2 polymorphisms were not associated with 17-AAG toxicity. The CYP3A5*3 polymorphism was associated with higher 17-AAG clearance.
Heat shock protein (Hsp) 90 is a cellular protein responsible for chaperoning multiple client proteins necessary for cell signaling, proliferation, and survival.1,2 Hsp90 is stress induced but is also constitutively expressed in unstressed cells. It forms a multimolecular complex that plays a regulatory role in the folding, intracellular transport, and repair or degradation of client proteins, many of which are protein kinases and transcription factors involved in signal transduction.3-5 Whitesell et al6 reported that geldanamycin (GA) bound to Hsp90, and this resulted in disruption of the Hsp90 function. The binding of GA causes a conformational change from the adenosine triphosphate–to the adenosine diphosphate–bound conformation of Hsp90, which cannot chaperone client proteins. Inhibition of Hsp90 function by GA results in dissociation of client proteins, such as HER2, RAF, mutant p53, CDK4, SRC, FAK, AKT, NFκB, and IGFR1,7-9 resulting in their degradation. Binding of GA and its analogs to Hsp90 also induces a stress response that is manifested, in part, by increased levels of cochaperone and other stress proteins such as Hsp70.10
Preclinical toxicology studies revealed that GA caused hepatotoxicity in dogs, so further evaluation was terminated.11 The GA analog 17-allylamino-17-demethoxygeldanamycin (17-AAG; NSC 330507) proved to be less toxic than GA and was active in mouse xenograft models.12-16 Preclinical pharmacology revealed that 17-AAG had excellent bioavailability when administered intraperitoneally but only modest oral bioavailability. It was metabolized to 17-aminogeldanamycin (17-AG) by cytochrome P450 (CYP) 3A4 and was widely distributed in body tissues but not the CNS.17,18 The metabolite 17-AG also binds to Hsp90, disrupting its ability to chaperone client proteins. Kelland et al12 also reported that 17-AAG metabolism by NAD(P)H:quinone oxidoreductase 1 (NQO1) increases its antitumor activity. The potential to target multiple signaling pathways for cancer cell proliferation and survival, its acceptable safety profile, and preclinical activity has led to the clinical evaluation of 17-AAG.
We designed a phase I dose-escalation study to evaluate the toxicities and determine the maximum-tolerated dose (MTD) of 17-AAG when administered as a weekly infusion (days 1, 8, and 15 of a 28-day cycle) and to determine the effect of 17-AAG on client and cochaperone proteins as possible biomarkers of target effect. If Hsp90 was affected by the treatment, we expected that the cochaperone proteins would increase and client proteins would decrease. To assess this, we collected peripheral-blood mononuclear cells (PBMCs) at different time points and immunoblotted for proteins involved in the chaperone complex and previously reported complex. This exploratory analysis was performed in a surrogate normal tissue to identify potential markers of drug effect that could be used in future trials. We also characterized the pharmacokinetics of 17-AAG and determined the effect of CYP3A5 and NQO1 polymorphisms on the disposition and toxicity of 17-AAG.
Patients with histologically or cytologically confirmed measurable or assessable metastatic or locally advanced cancer for which no established life-prolonging therapy was available were eligible for this study. Patients were ≥ 18 years, had a life expectancy of more than 12 weeks and an Eastern Cooperative Oncology Group performance status ≤ 2, and were willing to provide all biologic specimens as required by the protocol.
Patients were excluded if they had received chemotherapy, radiation therapy, biologic therapy, or immunotherapy ≤ 4 weeks before study registration (≤ 6 weeks in patients treated with mitomycin or nitrosoureas), if they had failed to recover from any toxic effects of prior treatment, or if they received radiation therapy to ≤ 25% of bone marrow. Patients were also excluded in the presence of the following: neutrophil count less than 1,500/μL; platelet count less than 100,000/μL; hemoglobin less than 9.0 g/dL; serum creatinine more than 1.25 × upper limit of normal (ULN) or actual or estimated creatinine clearance ≤ 60 mL/min (using the Cockcroft-Gault formula); total bilirubin more than 2 × ULN and AST more than 2.5 × ULN; and alkaline phosphatase more than 2 × ULN in the absence of demonstrable liver metastases or more than 5 × ULN in the presence of liver metastases. Other exclusion criteria included active or uncontrolled infection; concomitant pregnancy, lactation, or unwillingness to use adequate contraception; known CNS metastases or uncontrolled seizure disorder; uncontrolled intercurrent illness including, but not limited to, symptomatic congestive heart failure (New York Heart Association classification III or IV), unstable angina pectoris, or cardiac arrhythmia; and history of serious allergic reaction to eggs. Patients receiving known CYP450 3A4 inhibitors were excluded. All patients gave written informed consent according to institutional and federal guidelines.
17-AAG was supplied by National Cancer Institute as a sterile single-use amber vial containing 50 mg of 17-AAG in 2 mL of frozen dimethylsulfoxide. Egg phospholipids (EPL) diluent (NSC 704057) was supplied in a 50-mL flint vial containing 48 mL of 2% egg phospholipids and 5% dextrose in water for injection (US Pharmacopeia). After thawing the drug, the diluent was added to give a final concentration of 1 mg/mL, which was dispensed in a glass bottle.
Patients received 17-AAG intravenously as a 60-minute infusion on days 1, 8, and 15 of a 28-day cycle using a Simon accelerated titration design with intrapatient dose escalation. One patient per cohort was entered, and doses were increased by 40% increments. Patients were observed for a minimum of 4 weeks before new patients were treated, and dose escalation for the same patient was allowed if the toxicities of the previous cycle were grade 0. The accelerated phase ended when one patient experienced dose-limiting toxicities (DLT) during the first course or two patients experienced moderate toxicity (any grade 2 hematologic or nonhematologic toxicity).
Once the accelerated phase ended, standard dose escalation proceeded, whereby groups of three to six patients were treated at each dose level. If none of three patients experienced first-course DLT, three additional patients were treated at the next dose level (1.4 dose factor). If one of three patients experienced DLT, up to three more patients were entered at that same dose level. If two or more patients experienced DLT, no further patients were started at that dose, and three more patients were treated at the next lower dose level to more fully assess the toxicities associated with the MTD. The MTD was defined as one dose level below the dose that induced DLTs in one third or more of patients (at least two of a maximum of six new patients).
All toxicities were graded according to the National Cancer Institute Common Toxicity Criteria (version 2). MTD was defined based on toxicities documented in the first cycle of treatment only. Grade 3 or 4 nonhematologic toxicity (with the exception of nausea, vomiting, and diarrhea) was considered dose limiting. Grade 3 or 4 nausea, vomiting, or diarrhea in patients who had received prophylaxis and treatment with an optimal antiemetic or antidiarrheal regimen were considered dose limiting. Grade 4 neutropenia lasting ≥ 5 days or associated with fever or infection and grade 4 thrombocytopenia or anemia of any duration were also DLTs. Any omission of either the day 8 or 15 dose during cycle 1 was a DLT.
A complete patient history, examination, CBC, electrolytes, and chemistries were performed at baseline and before each course of treatment. CBCs were performed weekly during the study. Radiographic evaluation was performed at baseline and after every two courses of therapy to assess tumor response. A partial response (PR) was defined as a ≥ 50% reduction in the sum of the largest perpendicular diameters of indicator lesion(s), which had been chosen before therapy. A complete response (CR) was the disappearance of all evidence of tumor. For a patient to qualify for CR or PR, none of the factors constituting progression could be present. Tumor progression was defined as the appearance of a new lesion or lesions or a 25% increase in size of indicator lesion(s). Stable disease was documented when there was a failure to meet the criteria for CR, PR, or progression. All objective responses were required to last for at least 4 weeks.
Blood samples (5 mL) were drawn into heparin-containing tubes at the following times during cycle 1: before treatment, 0.5 and 0.9 hours after start of infusion, and 0.08, 0.25, 0.5, 1, 2, 4, 8, and 24 hours after the end of infusion. Blood samples were collected from a site remote from the 17-AAG infusion line and immediately cooled in ice water. Plasma was separated by centrifugation (1,000 to 1,200 × g for 10 minutes) at 4°C and transferred into plastic tubes that were stored at −70°C. Urine was collected for a 25-hour interval from the beginning of infusion to the time of the last blood sample.
Plasma and urine concentrations of 17-AAG and 17-AG were determined by the reverse-phase high-performance liquid chromatography procedure of Egorin et al.18 Separation of 17-AAG, 17-AG, and the internal standard alpha-naphthoflavone was achieved on a Genesis (Jones Chromatography, Lakewood, CO) ODS column (250 cm × 4.6 mm internal diameter, 4 μm) fitted with a Newguard RP-8 (Brownlee, distributed by Chrom Tech, Apple Valley, MN) precolumn (15 × 3.2 mm internal diameter, 7 μm). The mobile phase consisted of a mixture of 25 mmol/L potassium phosphate (pH 3):acetonitrile (volume-to-volume ratio, 45:55). The flow rate and detection wavelength were 1.0 mL/min and 334 nm, respectively. Plasma samples (1.0 mL) were extracted with 80:20 ethyl acetate:hexane (5.0 mL). After vigorous shaking for 15 minutes and centrifugation (1,000 × g) for 10 minutes, 4.0 mL of the organic layer evaporated to dryness under a gentle nitrogen stream. The residue was reconstituted in 200 μL mobile phase containing 10 μg/mL desipramine, and 100 μL was injected onto the high-performance liquid chromatography.
17-AAG and 17-AG plasma concentration data were analyzed by noncompartmental methods using the program WINNonlin (Scientific Consultant, Apex, NC). The apparent terminal elimination rate constants (kz) were determined by linear least-squares regression through the 5- to 25-hour plasma concentration–time points. The apparent elimination half-life (t1/2) was calculated as 0.693/kz. Areas under the plasma concentration–time curves (AUC) were determined using the linear trapezoidal rule from time zero to the time of the last detectable sample (Clast). AUCs through infinite time (AUC0-∞) were calculated by adding the value Clast/kz to AUClast. The clearance of 17-AAG was calculated as dose divided by AUC0-∞.
PBMCs were isolated before therapy, and DNA was extracted using a QIAamp DNA Mini kit (Qiagen, Valencia, CA) per instructions. CYP3A5*3 genotyping was performed using a modified method of Hustert et al.19 Specifically, M13 forward or reverse sequences were added to the 5′-terminus of each primer for use in dye primer sequencing chemistry. Amplicons were sequenced with an ABI 377 DNA sequencer (Applied Biosystems, Foster City, CA) using BigDye (Applied Biosystems) primer cycle sequencing chemistry. Samples were sequenced on both strands, and samples with ambiguous sequencing chromatograms were subjected to a second independent amplification, followed by DNA sequencing. The DNA sequence was visually analyzed using the Sequencher program (Gene Codes Corp, Ann Arbor, MI).
The NQO1*2 polymorphism was genotyped using a modified method of Gaedigk et al.20 Oligonucleotide primer pairs (CCTGAGGCCTCCTTATCAG and CAAAGAGGCTGCTTGGAGC) were designed based on the NQO1 gene (GenBank accession nos. MJ81596 through MJ81600). Amplification reactions used 2.5 units of AmpliTaq Gold DNA polymerase (Perkin Elmer, Foster City, CA) in a 50-μL reaction mixture containing 25 to 50 ng of DNA sample, 12.5 pmol of each primer, 0.05 mmol/L of deoxynucleotide triphosphates (Boehringer Mannheim, Indianapolis, IN), and 5 μL of 10 × polymerase chain reaction buffer containing 15 mmol/L of MgCl2 (Perkin Elmer). Polymerase chain reaction cycling parameters involved a 12-minute hot start at 94°C, followed by 35 cycles of 94°C for 30 seconds, 60°C for 30 seconds, and 72°C for 45 seconds, with a final 10 minutes extension at 72°C. Each 238-base pair amplicon was digested with 10 units of HinfI (New England Biolabs, Beverly, MA) and separated on a 3% agarose gel (Invitrogen, Carlsbad, CA). The homozygous *2 genotype produced DNA fragment sizes of 151 and 132 base pairs, whereas the homozygous wild type, *1, remained undigested.
Blood samples (35 mL) were drawn into heparin-containing tubes (Vacutainer CPT; Becton Dickinson, Franklin Lakes, NJ) before treatment and 6 and 25 hours after the beginning of treatment (days 1 and 15 of the treatment cycle). PBMCs were isolated, suspended in approximately 4 volumes of buffer (10 mmol/L Tris, 0.1 mmol/L EDTA, pH 7.5, plus protease inhibitors: 0.1 mmol/L leupeptin, 0.1 mg/mL bacitracin, 77 μg/mL aprotinin, 1.5 μmol/L pepstatin, and 1 mmol/L 4-[2-aminoethyl] benzenesulfanyl fluoride), and lysed by brief sonication. Total protein concentration was measured using Coomassie Plus reagent (Pierce, Rockford, IL).
Changes in the marker protein levels were measured by Western blotting. Cell lysate samples of known protein concentration (5 to 30 μg per sample) were resolved by sodium dodecyl sulfate- polyacrylamide gel electrophoresis and transferred to PVDF membranes (Millipore, Billerica, MA) for use in Western blotting.21 Blots were imaged with the Phase One Digital Imaging Program (Phase One Denmark A/S, Copenhagen, Denmark) and individual bands were analyzed with the IP LabGel Program (Signal Analytics, Vienna, VA).
The proteins detected and antibodies used were as follows: Hsp90, H901022; Hsp70, SPA 810 (StressGen; Stressgen Biotechnologies Corp, Victoria, BC, Canada); Hsf1, PA3-017 (Affinity BioReagents, Golden, CO); HDJ2, MS-225 (NeoMarkers; Lab Vision Corp, Freemont, CA); calcineurin, SC-9070 (Santa Cruz Biotechnology, Santa Cruz, CA); Hop, F523; Grp78, SC-1051 (Santa Cruz Biotechnology); FKBP51, H251a (gift from David Smith, Mayo Clinic, Scottsdale, AZ); FKBP52, EC1 (gift from Lee Faber, Medical College of Ohio, Toledo, OH); p23, JJ321; Cdc37, MA3-029 (Affinity BioReagents); RAF-A, SC-408 (Santa Cruz Biotechnology); RAF1, SC-133 (Santa Cruz Biotechnology); CHIP (gift from Avrom Caplan, Mt Sinai School of Medicine, New York, NY); and p56lck (gift from Steven Hartson, Oklahoma State University, Stillwater, OK).
Descriptive statistics form the basic analysis of this phase I study. An accelerated titration design was used, wherein cohorts of individual patients were accrued at successive dose levels until moderate toxicity was observed.24 Subsequently, a standard cohorts of three design was used.
Toxicity and response rates are presented using simple estimates of proportions. The relationships among the pharmacokinetics were explored using Pearson's correlation coefficients. Significance testing for the biomarker relationships to the 17-AAG pharmacokinetics was carried out by similar correlation analyses. All analyses were exploratory in nature. The sample size of 21 patients provided 80% power to detect population correlation coefficients of at least 0.58 using a two-sided alternative and a 5% type I error rate.
Twenty-one patients (Table 1) received 52 courses of treatment at 11 different dose levels (Table 2). The median age was 61 years (range, 38 to 77 years), and the median number of courses was two (range, one to eight courses). The most common tumor types treated included colorectal (n = 12) and lung (n = 3). Nineteen patients had a median of three prior chemotherapy regimens (range, zero to five regimens). At the MTD, eight patients were enrolled, but one received only one dose of 17-AAG. Pharmacokinetics were performed on the eight patients, whereas seven patients were assessable clinically.
Table 3 lists the treatment-related toxicities (cycle 1 and all cycles). Grade 3 DLT occurred in two patients at the 431 mg/m2 dose level; one patient had grade 3 hepatic enzyme (bilirubin and AST) elevation, and the other patient was hospitalized for grade 3 nausea and vomiting with associated anorexia and dehydration, despite adequate antiemetic treatment. The most common cycle 1 nonhematologic toxicities observed at the MTD (308 mg/m2) were diarrhea (n = 7), nausea (n = 6), vomiting (n = 6), fatigue (n = 6), anorexia (n = 5), and anemia (n = 2). At the MTD, one patient with pre-existing pulmonary infiltrates and prior oxaliplatin exposure died after developing grade 4 dyspnea and hypoxia 24 hours after receiving the first dose of 17-AAG. All cultures were negative, and a postmortem was not performed. Neither thrombocytopenia nor neutropenia was dose limiting.
All patients were assessable for response. No tumor responses were observed.
The pharmacokinetics of 17-AAG and 17-AG were studied in all patients during cycle 1. After the end of the 1-hour infusion, 17-AAG plasma concentration declined in a polyexponential manner, as illustrated for patients administered 308 mg/m2 of 17-AAG (Fig 1). Because a single compartmental model did not fit the plasma concentration–time data for all patients, data were fit by noncompartmental analysis. Pharmacokinetic parameter estimates are listed in Table 4. A statistically significant negative correlation (r = −0.7491, P < .0001) was observed between 17-AAG clearance and dose level (Fig 2), suggesting nonlinear pharmacokinetics. At the MTD (308 mg/m2), the mean t1/2 and clearance values for 17-AAG were 4.15 hours and 11.8 L/h/m2, respectively. The relationship between clearance and body-surface area was not significant (r = −0.01, P = .964).
The active metabolite, 17-AG, was detected at all dose levels. Peak plasma concentrations were observed 2.2 hours after beginning the 17-AAG infusion, and the mean t1/2 was 7.63 ± 8.63 hours at the MTD (Table 4). Patients treated with 17-AAG at the MTD (308 mg/m2) were able to achieve 17-AAG and 17-AG concentrations greater than 1 μmol/L for longer than 8 hours (Fig 1). Furthermore, because 17-AG can bind Hsp90 with comparable potency as 17-AAG, biologically active agents greater than 1 μmol/L were present in plasma for at least 24 hours at the recommended phase II dose. The 17-AG to 17-AAG AUC ratio was 1.01 (range, 0.21 to 2.45) and was unrelated to dose (data not shown). The recoveries of 17-AAG and 17-AG in the 24-hour urine were 3.08% (range, 0.58% to 6.96%) and 4.70% (range, 1.79% to 13.34%), respectively. No correlation between the 17-AAG and 17-AG peak plasma concentrations, t1/2, or AUC was observed.
Two patients carried the *1/*3 genotype, and the remaining 19 patients were homozygous for the CYP3A5*3 allele (q = 0.05). The pharmacokinetic and pharmacogenetic correlation was available on 20 patients. The two patients with the *1/*3 genotype had a higher mean clearance (30.7 L/h/m2) and higher mean 17-AG to 17-AAG ratio (2.13) compared with patients homozygous for the *3 allele (14.3 L/h/m2 and 0.90, respectively; Fig 3).
Twenty-one patients were genotyped for the NQO1*2 polymorphism. Fourteen were homozygous for the C/C genotype, five were heterozygous for the C/T genotype, and one was homozygous for the T/T genotype (q = 0.19). No correlation between 17-AAG disposition or toxicity and NQO1 genotype was observed (data not shown).
Hsp90 levels did not change significantly after drug treatment; however, in all patients, Hsp70 increased from 120% to 300% of control at both 6 and 25 hours after treatment but was not dose dependent (Figs. 4 and 5A). At day 15, this response was attenuated somewhat (Fig 5B). Less consistent and smaller increases were observed for the Hsp70 homolog in the endoplasmic reticulum, Grp78, and for another cochaperone, the human Hsp40 protein, HDJ2. The Hsp90 cochaperones Hop, Cdc37, CHIP, FKBP51, FKBP52, and p23 were unchanged.
We tested PBMC protein lysates for multiple Hsp90 client proteins, including protein kinase p56lck, insulin-like growth factor receptor 1, CDK4, RAF-A, RAF1, Hsf1, Ah receptor, glucocorticoid receptor, and calcineurin A. Inconsistent decreases in the level of calcineurin A were occasionally observed; however, the remaining client proteins were unchanged. No correlation between change in chaperone or client protein levels and 17-AAG and 17-AG pharmacokinetic parameters was seen (data not shown).
A growing understanding of the molecular, genetic, and biochemical changes that occur during the process of carcinogenesis, progression, and metastasis has shifted the focus of drug development from empiric therapy towards therapeutics that act on specific molecular targets responsible for the malignant phenotype.25 However, because of the redundancy in the pathways for tumorigenesis and the multiple interactive signaling routes, significant obstacles exist for the development of effective targeted approaches in solid tumors. 17-AAG represents a class of drugs (benzoquinone ansamycin antibiotics) capable of affecting multiple proteins in signal transduction pathways involved in tumor-cell proliferation and survival and is the first drug of this class to reach clinical trials.
In the current study, the MTD and recommended phase II dose of 17-AAG was 308 mg/m2. Hepatic toxicity that had precluded the development of the parent drug, GA, was not observed at the MTD. At the 431 mg/m2 dose level, we observed a transient increase in AST and bilirubin in a patient with hepatoma. This may have reflected the limited hepatic reserve of the patient. Nevertheless, he met the criteria of a DLT as defined by protocol. The second patient at this dose level had DLTs consisting of nausea, vomiting, anorexia, and dehydration. These toxicities have been observed in other trials using the same schedule or different schedules of 17-AAG.26
The major findings of the pharmacokinetic investigations of 17-AAG were as follows: biologically active plasma concentrations of 17-AAG and 17-AG, both of which bind Hsp90, could be achieved at the MTD; 17-AAG may exhibit nonlinear pharmacokinetics; there is no relationship between 17-AAG dose and body-surface area; and there is no relationship between pharmacokinetic parameters and toxicity or biomarker effect. The combined plasma concentrations of 17-AAG and 17-AG of more than 1 μmol/L for 24 hours achieved at the MTD is effective in causing client degradation in preclinical studies. This, together with the observation that intracellular concentrations of 17-AAG and 17-AG are greater than whole-tissue levels of 17-AAG and 17-AG,27 suggest that biologically relevant concentrations are achieved at the MTD.
The plasma clearance of 17-AAG seemed to decrease as the dose was increased by nearly 30-fold from 15 mg/m2 to 431 mg/m2 (Table 4), which is suggestive of nonlinear pharmacokinetics. However, the elimination t1/2 increased only three-fold from 1.34 ± 0.51 hours for patients receiving the lowest doses (15 to 57 mg/m2) to 4.12 ± 1.76 hours for patients receiving the higher doses (80 to 431 mg/m2), whereas the 17-AG/17-AAG AUC ratios remained constant (1.26 ± 1.01 v 0.96 ± 0.45). The longer t1/2 values may be better estimates of the elimination t1/2 at higher doses because the drug concentrations remain above the assay detection limit for a longer time. At the lower dose levels, this could result in an underestimate of the AUC values, a concomitant overestimate of clearance, and a suggestion of nonlinear pharmacokinetics. Because this trial design with single patients at individual dose levels limits the potential to evaluate the reproducibility of this observation, the results from other trials using a similar schedule and dose-escalation scheme will be important to validate these findings.
The absence of a relationship between 17-AAG clearance and body-surface area for patients enrolled onto this study suggests that a fixed dose may be administered to adults in phase II studies. On the basis of tolerability of 17-AAG in patients administered the 308 mg/m2 dose, a fixed dose of 620 mg should be tolerable. However, because this trial design incorporated dosing adjusted for body-surface area and 17-AAG clearance varied over a five-fold range (6.0 to 31.5 L/h) for the seven patients treated at the MTD, whereas the body-surface area in those same patients fell in a narrow range (1.8 to 2.3 m2), this fixed dose should be confirmed before it is used in a phase II study. Furthermore, the body-surface area of all the patients enrolled onto this trial also fell into a narrow range (1.58 to 2.96 m2). A fixed dose defined for an adult population cannot be extrapolated to the pediatric population, which generally exhibits a broader range and includes patients with much smaller body-surface area values.
We found no relationship between 17-AAG pharmacokinetics and its clinical toxicity. This is not surprising because the low frequency of toxicity and the small sample size makes it unlikely that we could detect such an association. The absence of a correlation between 17-AAG pharmacokinetics and change in Hsp70 levels in PBMCs may similarly be a reflection of the small sample size. Furthermore, it is not possible to distinguish the contribution of 17-AAG and 17-AG to the effect on Hsp70, so the combined effect on Hsp70 cannot be correlated with the pharmacokinetics of 17-AAG or 17-AG separately.
We examined the effect of 17-AAG treatment in protein lysates from PBMCs because of ease of access and uncertainty of which proteins would be detectable in human tissue samples. Immunoblotting for proteins involved in the chaperone complex (Hsp90, Hsp70, Grp78, Hop, Cdc37, CHIP, FKBP51, FKBP52, and p23) was performed. Levels of Hsp90 did not change constitutively (Fig 5A). Whereas the inducible form of Hsp90 may have increased with 17-AAG treatment, the large quantity of constitutively expressed Hsp90 may mask any change that can be detected by the antibody that binds both forms of the protein. The increase in Hsp70 (Figs. 4 and 5A) that we observed is consistent with experimental results in many tumor models14,28-30 and clearly indicates that 17-AAG has affected the target in this normal tissue. Furthermore, it confirms that the concentrations of 17-AAG and 17-AG achieved in plasma are biologically active. However, we cannot discern whether the parent drug or metabolite is most important in causing this effect. The other proteins that participate in the chaperone complex did not change with 17-AAG treatment. The client proteins that were evaluated were selected based on reported relationships to Hsp90. We expected to detect a degradation of client proteins after 17-AAG treatment.14,31 Some proteins, such as RAF-B and AKT, were not detectable in PBMCs and could not be evaluated. None of the studied client proteins demonstrated consistent decreases in protein levels after 17-AAG treatment. This does not rule out a biologic effect on client proteins. Normal tissues may have the capacity to compensate for the effect of Hsp90 disruption, as suggested by Kamal et al.32 This would be consistent with the lack of myelosuppression that was seen with 17-AAG treatment. Alternatively, the timing of samples obtained may not have been optimal to identify the client protein degradation. An alternative approach to assessing a biologic effect may be to measure human epidermal growth factor receptor 2 ectodomain and insulin-like growth factor binding protein 2 in plasma, as suggested by Zhang et al.33 These markers of apoptosis were increased in murine xenografts treated with 17-AAG. The biomarker studies suggest that Hsp70 is a good indicator that Hsp90 has been affected but that effects on client proteins will need to be tested in tumor samples to better determine whether Hsp90 disruption leads to degradation of these proteins.
CYP3A5*3 polymorphism may substantially contribute to the interindividual differences in CYP3A-dependent drug clearance.19,34 We have found that the rate of 17-AAG metabolism by CYP3A4 and CYP3A5 was the same (unpublished observations). In our patient population, the two patients with the *1/*3 genotype had a higher clearance of 17-AAG and higher 17-AG/17-AAG AUC ratios than patients with *3/*3 genotype. This suggests that the CYP3A5*3 polymorphism may alter exposure to the parent drug, but is unlikely to have any clinical consequence, as 17-AG, the product of 17-AAG metabolism, is bioactive.35
Kelland et al12 suggested that NQO1 can increase tumor-cell sensitivity to 17-AAG. Six patients treated in this study had the C/T or T/T genotype, which are associated with decreased NQO1 enzyme activity. However, there was no difference in drug clearance or toxicity in these patients when compared with those with the homozygous C/C genotype that is associated with normal enzyme activity. This would suggest that NQO1 polymorphisms are unlikely to determine toxicity of 17-AAG.
This phase I study demonstrates that weekly (3 of 4 weeks) 17-AAG can be administered safely at a dose of 308 mg/m2. Combined plasma levels of 17-AAG and 17-AG are above biologically effective concentrations for more than 24 hours at this dose. Furthermore, we have demonstrated that the drug has a target effect (increases in Hsp70) in PBMCs. Ongoing studies of 17-AAG in an alternate schedule and in combination with chemotherapeutic agents will more clearly define the effect of Hsp90-targeted therapy on client proteins and chemosensitivity.7,36
The following authors or their immediate family members have indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. Consultant/Advisory Role: Charles Erlichman, Kosan. Honoraria: David Toft, Chiron, Syrrx. For a detailed description of these categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section of Information for Contributors found in the front of every issue.
Fig 1. Mean 17-allylamino-17-demethoxygeldanamycin (17-AAG) and 17-aminogeldanamycin (17-AG) plasma concentration versus time profiles for patients (n = 7) administered 17-AAG at the maximum-tolerated dose (308 mg/m2).
Fig 2. Relationship between 17-allylamino-17-demethoxygeldanamycin plasma clearance and dose level.
Fig 3. Comparison of the 17-aminogeldanamycin to 17-allylamino-17-demethoxygeldanamycin area under the curve (AUC) ratios for patients with the CYP3A5*1/*3 genotype and patients with the CYP3A5*3/*3 genotype. Shaded area contains the mean AUC ratio and two standard deviations (Std Dev) from the mean for patients with the *3/*3 genotype.
Fig 4. Western blots demonstrating heat shock protein (Hsp) 70 induction (0, 6, and 25 hours) for patients treated at the maximum-tolerated dose (308 mg/m2).
Fig 5. (A) Heat shock protein (Hsp) 90 and Hsp70 levels 25 hours after 17-AAG infusion. (B) Comparison of Hsp70 levels 25 hours after day 1 dose and 25 hours after day 15 dose. (A and B) One hundred percent represents the level of the protein in PBMCs obtained before that dose.
|
| Characteristic | No. of Patients |
|---|---|
| Age, years | |
| Median | 61 |
| Range | 38-77 |
| Sex | |
| Male | 10 |
| Female | 11 |
| PS | |
| 0 | 13 |
| 1 | 8 |
| Tumor type | |
| Colorectal | 12 |
| Lung | 3 |
| Small bowel | 1 |
| Liver | 1 |
| Anal | 1 |
| Ovary | 1 |
| Thyroid | 1 |
| Unknown | 1 |
| Prior therapy | |
| Surgery | 21 |
| Radiation therapy | 8 |
| Chemotherapy | 19 |
| Immunotherapy | 5 |
| Hormonal | 1 |
Abbreviation: PS, performance status.
|
| 17-AAG (mg/m2) | No. of Patients | No. of Courses | ||
|---|---|---|---|---|
| Median | Range | |||
| 15 | 1 | 2 | — | |
| 21 | 1 | 2 | — | |
| 29 | 1 | 4 | — | |
| 41 | 1 | 3 | — | |
| 57 | 1 | 2 | — | |
| 80 | 1 | 2 | — | |
| 112 | 1 | 2 | — | |
| 157 | 3 | 2 | 1-8 | |
| 200 | 1 | 1 | — | |
| 308 | 7 | 2 | 1-4 | |
| 431 | 3 | 1 | 1-8 | |
Abbreviation: 17-AAG, 17-allylamino-17-demethoxygeldanamycin.
|
| Grade | Toxicity (No. of patients) | ||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Anemia | Anorexia | Bilirubin | Dehydration | Diarrhea | Dyspnea | Fatigue | Hypoxia | Myalgia | Nausea | AST | Vomiting | ||||||||||||
| Cycle 1 only | |||||||||||||||||||||||
| 1 | 2 | 2 | 0 | 0 | 6 | 0 | 2 | 0 | 0 | 3 | 0 | 2 | |||||||||||
| 2 | 1 | 5 | 0 | 0 | 1 | 0 | 3 | 0 | 0 | 2 | 0 | 2 | |||||||||||
| 3 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 1 | 1 | 1 | 2 | |||||||||||
| 4 | 0 | 0 | 0 | 0 | 0 | 1* | 0 | 1* | 0 | 0 | 0 | 0 | |||||||||||
| All cycles | |||||||||||||||||||||||
| 1 | 2 | 4 | 0 | 1 | 7 | 0 | 2 | 0 | 0 | 4 | 0 | 1 | |||||||||||
| 2 | 1 | 6 | 0 | 0 | 2 | 0 | 3 | 0 | 0 | 3 | 0 | 2 | |||||||||||
| 3 | 1 | 1 | 1 | 1 | 1 | 1 | 1 | 0 | 1 | 2 | 1 | 3 | |||||||||||
| 4 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | |||||||||||
Abbreviation: NCICTC, National Cancer Institute Common Toxicity Criteria.
*One death as a result of respiratory failure.
|
| Dose (mg/m2) | No. of Patients | 17-AAG | 17-AG | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Cmax (μg/ml) | t1/2z (hours) | AUC (μg/mL/h) | CI (L/h) | CI (L/h/m2) | Vss (L/m2) | tmax (hours) | Cmax (μg/ml) | t1/2z (hours) | AUC (μg/mL/h) | ||||||||||
| 15 | 1 | 0.300 | 1.46 | 0.564 | 56.3 | 26.8 | 46.4 | 1.50 | 0.055 | 2.37 | 0.203 | ||||||||
| 21 | 1 | 0.315 | 1.90 | 0.509 | 94.9 | 41.3 | 102 | 1.50 | 0.093 | 19.0 | 1.24 | ||||||||
| 29 | 1 | 0.525 | 1.31 | 1.08 | 58.8 | 26.7 | 75.3 | 1.25 | 0.101 | 15.7 | 1.89 | ||||||||
| 41 | 1 | 0.770 | 0.68 | 1.64 | 39.8 | 25.2 | 67.9 | 5.00 | 0.090 | 4.51 | 0.797 | ||||||||
| 57 | 1 | 1.41 | 5.55 | 3.65 | 31.3 | 15.8 | 87.1 | 3.00 | 0.690 | 11.5 | 4.34 | ||||||||
| 80 | 1 | 1.53 | 3.93 | 2.87 | 62.4 | 28.1 | 84.9 | 0.92 | 0.306 | 7.65 | 1.59 | ||||||||
| 112 | 1 | 2.74 | 4.31 | 5.79 | 38.7 | 18.4 | 62.3 | 1.25 | 0.604 | 8.10 | 4.9 | ||||||||
| 157 | 3 | ||||||||||||||||||
| Mean | 4.70 | 4.09 | 10.2 | 36.0 | 18.3 | 63.7 | 1.44 | 1.96 | 7.34 | 12.7 | |||||||||
| SD | 1.05 | 1.74 | 3.3 | 13.0 | 5.9 | 40.6 | 0.40 | 0.91 | 1.47 | 5.3 | |||||||||
| 200 | 1 | 5.30 | 1.55 | 14.5 | 26.0 | 13.1 | 29.9 | 2.00 | 2.00 | 1.95 | 10.0 | ||||||||
| 308 | 8 | ||||||||||||||||||
| Mean | 9.09 | 4.15 | 33.2 | 23.4 | 11.6 | 42.8 | 2.57 | 2.26 | 7.63 | 37.0 | |||||||||
| SD | 2.37 | 1.98 | 25.1 | 7.9 | 4.0 | 13.7 | 0.64 | 0.69 | 8.63 | 45.8 | |||||||||
| 431 | 3 | ||||||||||||||||||
| Mean | 15.1 | 4.47 | 55.4 | 21.0 | 10.8 | 40.6 | 2.25 | 4.28 | 7.20 | 41.1 | |||||||||
| SD | 2.8 | 1.48 | 29.5 | 11.8 | 7.5 | 13.1 | 0.74 | 3.46 | 3.49 | 34.3 | |||||||||
Abbreviations: 17-AAG, 17-allylamino-17-demethoxygeldanamycin; 17-AG, 17-aminogeldanamycin; Cmax, maximum concentration; t1/2z, elimination half-life; AUC, area under the curve; Cl, clearance; Vss, steady state volume of distribution; tmax, time to maximum serum concentration; SD, standard deviation.
Supported in part by grant Nos. CA15083, CA90390, and CA69912 (National Cancer Institute), and M01-RR00585 (NCRR).
Authors' disclosures of potential conflicts of interest are found at the end of this article.
We thank Sacha Nelson (Protocol Development Coordinator), Debra Sprau (Certified Research Associate), and the patients who participated in this trial.
| 1. | Morimoto RI, Kline MP, Bimston DN, et al: The heat-shock response: Regulation and function of heat-shock proteins and molecular chaperones. Essays Biochem 32::17,1997-29, Medline, Google Scholar |
| 2. | Hartl FU, Hayer-Hartl M: Molecular chaperones in the cytosol: From nascent chain to folded protein. Science 295::1852,2002-1858, Crossref, Medline, Google Scholar |
| 3. | Young JC, Moarefi I, Hartl FU: Hsp90: A specialized but essential protein-folding tool. J Cell Biol 154::267,2001-273, Crossref, Medline, Google Scholar |
| 4. | Pratt WB, Toft DO: Steroid receptor interactions with heat shock protein and immunophilin chaperones. Endocr Rev 18::306,1997-360, Medline, Google Scholar |
| 5. | Richter K, Buchner J: Hsp90: Chaperoning signal transduction. J Cell Physiol 188::281,2001-290, Crossref, Medline, Google Scholar |
| 6. | Whitesell L, Mimnaugh EG, De Costa B, et al: Inhibition of heat shock protein HSP90-pp60v-src heteroprotein complex formation by benzoquinone ansamycins: Essential role for stress proteins in oncogenic transformation. Proc Natl Acad Sci U S A 91::8324,1994-8328, Crossref, Medline, Google Scholar |
| 7. | Goetz MP, Toft DO, Ames MM, et al: The Hsp90 chaperone complex as a novel target for cancer therapy. Ann Oncol 14::1169,2003-1176, Crossref, Medline, Google Scholar |
| 8. | Neckers L: Hsp90 inhibitors as novel cancer chemotherapeutic agents. Trends Mol Med 8::S55,2002-S61, Medline, Google Scholar |
| 9. | Workman P: Overview: Translating Hsp90 biology into Hsp90 drugs. Curr Cancer Drug Targets 3::297,2003-300, Crossref, Medline, Google Scholar |
| 10. | Whitesell L, Bagatell R, Falsey R: The stress response: Implications for the clinical development of hsp90 inhibitors. Curr Cancer Drug Targets 3::349,2003-358, Crossref, Medline, Google Scholar |
| 11. | Supko JG, Hickman RL, Grever MR, et al: Preclinical pharmacologic evaluation of geldanamycin as an antitumor agent. Cancer Chemother Pharmacol 36::305,1995-315, Crossref, Medline, Google Scholar |
| 12. | Kelland LR, Sharp SY, Rogers PM, et al: DT-Diaphorase expression and tumor cell sensitivity to 17-allylamino, 17-demethoxygeldanamycin, an inhibitor of heat shock protein 90. J Natl Cancer Inst 91::1940,1999-1949, Crossref, Medline, Google Scholar |
| 13. | Burger AM, Fiebig HH, Stinson SF, et al: 17-(Allylamino)-17-demethoxygeldanamycin activity in human melanoma models. Anticancer Drugs 15::377,2004-387, Crossref, Medline, Google Scholar |
| 14. | Bagatell R, Khan O, Paine-Murrieta G, et al: Destabilization of steroid receptors by heat shock protein 90-binding drugs: A ligand-independent approach to hormonal therapy of breast cancer. Clin Cancer Res 7::2076,2001-2084, Medline, Google Scholar |
| 15. | Nguyen DM, Lorang D, Chen GA, et al: Enhancement of paclitaxel-mediated cytotoxicity in lung cancer cells by 17-allylamino geldanamycin: In vitro and in vivo analysis. Ann Thorac Surg 72::371,2001-378, Crossref, Medline, Google Scholar |
| 16. | Yang J, Yang JM, Iannone M, et al: Disruption of the EF-2 kinase/Hsp90 protein complex: A possible mechanism to inhibit glioblastoma by geldanamycin. Cancer Res 61::4010,2001-4016, Medline, Google Scholar |
| 17. | Egorin MJ, Zuhowski EG, Rosen DM, et al: Plasma pharmacokinetics and tissue distribution of 17-(allylamino)-17-demethoxygeldanamycin (NSC 330507) in CD2F1 mice1. Cancer Chemother Pharmacol 47::291,2001-302, Crossref, Medline, Google Scholar |
| 18. | Egorin MJ, Rosen DM, Wolff JH, et al: Metabolism of 17-(allylamino)-17-demethoxygeldanamycin (NSC 330507) by murine and human hepatic preparations. Cancer Res 58::2385,1998-2396, Medline, Google Scholar |
| 19. | Hustert E, Haberl M, Burk O, et al: The genetic determinants of the CYP3A5 polymorphism. Pharmacogenetics 11::773,2001-779, Crossref, Medline, Google Scholar |
| 20. | Gaedigk A, Tyndale RF, Jurima-Romet M, et al: NAD(P)H: quinone oxidoreductase: Polymorphisms and allele frequencies in Caucasian, Chinese and Canadian Native Indian and Inuit populations. Pharmacogenetics 8::305,1998-313, Crossref, Medline, Google Scholar |
| 21. | Johnson JL, Toft DO: Binding of p23 and hsp90 during assembly with the progesterone receptor. Mol Endocrinol 9::670,1995-678, Medline, Google Scholar |
| 22. | Barent RL, Nair SC, Carr DC, et al: Analysis of FKBP51/FKBP52 chimeras and mutants for Hsp90 binding and association with progesterone receptor complexes. Mol Endocrinol 12::342,1998-354, Crossref, Medline, Google Scholar |
| 23. | Smith DF, Sullivan WP, Marion TN, et al: Identification of a 60-kilodalton stress-related protein, p60, which interacts with hsp90 and hsp70. Mol Cell Biol 13::869,1993-876, Crossref, Medline, Google Scholar |
| 24. | Simon R, Freidlin B, Rubinstein L, et al: Accelerated titration designs for phase I clinical trials in oncology. J Natl Cancer Inst 89::1138,1997-1147, Crossref, Medline, Google Scholar |
| 25. | Gelmon KA, Eisenhauer EA, Harris AL, et al: Anticancer agents targeting signaling molecules and cancer cell environment: Challenges for drug development? J Natl Cancer Inst 91::1281,1999-1287, Crossref, Medline, Google Scholar |
| 26. | Sausville EA, Tomaszewski JE, Ivy P: Clinical development of 17-allylamino, 17-demethoxygeldanamycin. Curr Cancer Drug Targets 3::377,2003-383, Crossref, Medline, Google Scholar |
| 27. | Xu L, Eiseman JL, Egorin MJ, et al: Physiologically-based pharmacokinetics and molecular pharmacodynamics of 17-(allylamino)-17-demethoxygeldanamycin and its active metabolite in tumor-bearing mice. J Pharmacokinet Pharmacodyn 30::185,2003-219, Crossref, Medline, Google Scholar |
| 28. | Lawson B, Brewer JW, Hendershot LM: Geldanamycin, an hsp90/GRP94-binding drug, induces increased transcription of endoplasmic reticulum (ER) chaperones via the ER stress pathway. J Cell Physiol 174::170,1998-178, Crossref, Medline, Google Scholar |
| 29. | Hegde RS, Zuo J, Voellmy R, et al: Short circuiting stress protein expression via a tyrosine kinase inhibitor, herbimycin A. J Cell Physiol 165::186,1995-200, Crossref, Medline, Google Scholar |
| 30. | Murakami Y, Uehara Y, Yamamoto C, et al: Induction of hsp 72/73 by herbimycin A, an inhibitor of transformation by tyrosine kinase oncogenes. Exp Cell Res 195::338,1991-344, Crossref, Medline, Google Scholar |
| 31. | Schneider C, Sepp-Lorenzino L, Nimmesgern E, et al: Pharmacologic shifting of a balance between protein refolding and degradation mediated by Hsp90. Proc Natl Acad Sci U S A 93::14536,1996-14541, Crossref, Medline, Google Scholar |
| 32. | Kamal A, Thao L, Sensintaffar J, et al: A high-affinity conformation of Hsp90 confers tumour selectivity on Hsp90 inhibitors. Nature 425::407,2003-410, Crossref, Medline, Google Scholar |
| 33. | Zhang HJ, Chung D, Yong-Ching Y, et al: Identification of new biomarkers for clinical trials of hsp90 inhibitors. Proc Am Assoc Cancer Res 45::894,2004, (abstr 3877) Google Scholar |
| 34. | Kuehl P, Zhang J, Lin Y, et al: Sequence diversity in CYP3A promoters and characterization of the genetic basis of polymorphic CYP3A5 expression. Nat Genet 27::383,2001-391, Crossref, Medline, Google Scholar |
| 35. | Schnur RC, Corman ML, Gallaschun RJ, et al: Inhibition of the oncogene product p185erbB-2 in vitro and in vivo by geldanamycin and dihydrogeldanamycin derivatives. J Med Chem 38::3806,1995-3812, Crossref, Medline, Google Scholar |
| 36. | Munster PN, Basso A, Solit D, et al: Modulation of Hsp90 function by ansamycins sensitizes breast cancer cells to chemotherapy-induced apoptosis in an RB- and schedule-dependent manner. See: E. A. Sausville, Combining cytotoxics and 17-allylamino, 17-demethoxygeldanamycin: Sequence and tumor biology matters, Clin Cancer Res 7:2155-2158, 2001. Clin Cancer Res 7::2228,2001-2236, Medline, Google Scholar |
