2003 AACR-NCI-EORTC epothilone abstracts:
A248 Phase 1 dose escalating and pharmacokinetic (PK) study of kos-862 (epothilone D): Phase 2 dose and schedule defined.
David Spriggs, Jakob Dupont, Sandy Pezzulli, Joseph Larkin, Gillian Cropp, Robert Johnson, and Alison L. Hannah. Memorial Sloan-Kettering Cancer Center, New York, NY and Kosan Biosciences, Inc., Hayward, CA.
KOS-862 (Epothilone D) is a naturally occurring cytotoxic macrolide that stabilizes microtubules, inducing mitotic arrest. Preclinical data suggest KOS-862 has comparable activity to paclitaxel, but retains efficacy in MDR overexpressing cells. Previous Phase 1 schedules showed drug-related toxicities recovering within 7 days following an IV infusion. Protocol objectives: define the Phase 2 dose and schedule of KOS-862, investigating dose intense schedules: 3 weeks out of every 4 and 2 weeks out of every 3. Patients had a diagnosis of advanced solid tumors or lymphoma.
Groups of 3 pts were treated at escalating doses of KOS-862. Toxicity was assessed in 3- or 4-week cycles. PK sampling occurred after the 1st and 3rd dose (plus after the 4th dose on the 4-week schedule); PD after the 1st dose at doses >/=100 mg/m2. Drug concentrations were assayed using LC/MS/MS (LLQ 2 ng/ml); PK analyzed using non-compartmental methods. PD was assessed by microtubule bundle formation in PBMCs by immunocytochemistry. 29 patients were enrolled (5 dose levels: 16-120 mg/m2); 100 mg/m2 was expanded on both schedules. Baseline demographics: median age 58 (25-77); median KPS 80 (70-100); 15 male; diagnoses: 8 ovarian, 5 colorectal, 4 lymphoma, 4 NSCLC; 8 other. DLT (one episode each of NVD/dehydration and brief visual hallucination) was observed at the highest dose. Drug-related toxicities (mild-to-moderate severity) at 100 mg/m2 included (n=10; 25+ cycles): fatigue (n=5), sensory neuropathy (n=5) and N/V/D (n=3). Although cognitive/perceptual changes comprise the DLT of this agent, other than mild dizziness (n=3) these were not observed at 100 mg/m2. Although the mean number of cycles at this dose was 2.5, sensory neurological toxicities did not appear to be cumulative, and were responsive to dose reduction (to 75 mg/m2) or dose delay (7 days). Other than the 2 pts with DLT, there were no withdrawals for drug-related toxicities. PK (n= 26; 15 at 100 mg/m2; 57 infusions) showed increases in mean Cmax (626, 1624, 2215, and 4037 ng/ml) and AUCtot (3088, 4610, 7752 and 12093 ng*h/ml) linear across 16-100 mg/m2. At this dose, there was no significant change in AUC comparing the 3 sampling days (p=0.37 by repeated ANOVA testing). PK analysis (mean±sd) for 100 mg/m2: half-life= 9.1 ± 2.2 hours; vz = 119 ± 41 l/m2; cl = 9.3 ± 3.2 l/h/m2; no dose dependency was observed for these parameters. Compared to less dose intense schedules, PK on this schedule maintains the systemic exposure with a slightly higher Cmax; other parameters are similar (although clearance showed a trend towards higher values on this schedule). Microtubule bundle formation in PBMCs (n=9) showed maximal values at the end-of-infusion and 1-hour post-infusion (~40% of PBMCs showed bundle formation), sustained at 15% 24-hours following the infusion. Tumor shrinkage was observed in 2 patients with large cell and mediastinal B-cell lymphoma. Stable disease (>3 months) was seen in renal, ovarian and mesothelioma; tumor marker declines (colorectal, ovarian) were observed. KOS-862 is a promising new agent; a dose of 100 mg/m2 is the recommended phase 2 dose. Phase 2 single-agent trials and combination studies using this schedule are planned.
Copyright © 2003 American Association for Cancer Research. Published as a Supplement to Clinical Cancer Research, Volume 9, Issue 16 (December 1, 2003).
A261 A phase I study to determine the safety and pharmacokinetics of continuous intravenous administration of KOS-862 (Epothilone D) in patients with solid tumors.
Kyle Holen, Alison Hannah, Yiqing Zhou, Gillian Cropp, Robert Johnson, Jennifer Volkman, Kim Binger, Dona Alberti, and George Wilding. University of Wisconsin, Madison, WI and Kosan Biosciences, Hayward, CA.
KOS-862 (Epothilone D) is a potent antitumor agent that enhances the assembly of microtubules and their stabilization against depolymerization in a manner similar to paclitaxel. KOS-862 has been shown to be more potent and less toxic than paclitaxel in multiple different mouse tumor models, particularly those derived from multidrug-resistant cells that over-express p-glycoprotein. Six schedules of KOS-862 are in phase 1 development. As the optimal schedule in the mouse xenograft model used prolonged exposure (6-hour IV infusion at 30 mg/kg), two continuous infusion (CI) schedules were attempted in addition to shorter IV schedules in phase 1 studies. This is the first description of KOS-862 administered as a CI over 24 - 72 hours every 2 weeks. In schedule A (24 hr CI every 2 weeks), the amount of mgs infused per hour increased in an accelerated fashion (starting at 1 mg/hr) until grade 2 toxicity was noted. In schedule B, 1 mg/hour CI was delivered for 48 hours prior to increasing the infusion time to 72 hours; dose escalation in mg/hr will be continued using the 72-hr CI. Computer simulations predicted that a loading dose (75 mg/hr X 30 minutes) would be required in order to sustain drug concentration. Ten patients have been enrolled (median age 67, range 54-76): 7 schedule A, 3 schedule B. A total of 24 infusions have been administered. Following one grade 3 hypersensitivity reaction, steroid and antihistamine premedications were used. Tumor types: 3 prostate, 2 colorectal, lung, esophageal, renal, bile duct and unknown primary (1 each). Schedule A drug-related toxicity included mild-to-moderate fatigue (n=2), nausea/anorexia/abdominal pain (n=1 each). Schedule B toxicities include grade 2 fatigue and grade 2 dizziness (n=1 each). DLT was not observed (in shorter infusional studies, this consisted of neurological toxicity, including abnormal gait and cognitive/perceptual changes at doses >/= 150 mg/m2). Neurological toxicity in this study was not notable (grade 1 sensory neuropathy n=1). Preliminary results of pharmacokinetic (PK) studies (n=7; 13 infusions) comparing two dose groups are presented below. Elimination half-life (9.8 + 2.0 hours) was within the same range as shorter infusion studies, as was Vz (118 ± 61 L/m2). Interday AUCtot values support stationary PK at this dosing interval; the exposure after 24-hour infusion at this low dose is encouraging. Css in the mouse xenograft model using CI equaled approximately 500 ng/ml (protein binding is higher in humans at 99% compared to 96%). As a pharmacodynamic (PD) marker of biological activity, PBMCs were examined for microtubule bundle formation via immunohistochemistry. Bundle formation in PBMCs was observed but at low levels (consistent with the doses administered). Comprehensive data on safety, tolerability, PKs and PDs will be presented.
PK Parameter 2 mg/hr; 24-hr CI 1 mg/hr; 48-hr CI
CL (L/hr/m2) 9.2 + 5.0 8.7 + 1.4
Cmax (ng/mL) 1487 ± 514 1077 ± 208
Cmin(ng/mL) 110 ± 40 45 ± 20
AUCtot (ng*h/mL) 6243 + 2802 4288 + 1313
AUCtot (Day1 vs Day15) +1.0%, -12%,11% +14%, -11%
Copyright © 2003 American Association for Cancer Research. Published as a Supplement to Clinical Cancer Research, Volume 9, Issue 16 (December 1, 2003).
A262 A comparative study of two novel, non-taxane microtubule stabilizing agents (XAA296A and EPO906A) utilizing the Extreme Drug Resistance (EDR) Assay in paclitaxel-sensitive or paclitaxel-resistant breast and ovarian cancers.
Monil Shah, Lisa Dauffenbach, Christopher A. Kerfoot, Pamela Cohen, Kenneth W. Culver, Sunil Sharma, and Markus Wartmann. Novartis Pharmaceuticals Corp., East Hanover, NJ, Oncotech, Inc., Tustin, CA, and Novartis Pharma, Basel, Switzerland.
Introduction: The Oncotech EDR® Assay, an in vitro drug resistance assay, has been shown to be a reliable test to predict resistance to anti-cancer drugs. In breast cancer treated with chemotherapy, the EDR status has been shown to be significantly associated with time to tumor progression and overall survival. Similarly, ovarian cancer patients with tumors classified as EDR to cisplatin have been shown to be at a significantly increased risk for progression and death when treated with standard platinum-based regimens. The EDR Assay was used to compare the effects of paclitaxel to XAA296A and EPO906A, two potent, novel non-taxane microtubule-stabilizing agents currently in clinical trials. Methods: The EDR Assay was performed utilizing disaggregated cells from fresh clinical tumor explants grown in soft agar. A panel of tumors (10 each of paclitaxel-sensitive or resistant ovarian cancer, and 10 each of paclitaxel-sensitive or resistant breast cancer) were utilized in this study. Tumor explants were exposed to high, moderate, and low concentrations of paclitaxel (1000, 10 and 0.1 nM), XAA296A (1000, 10 and 0.1 nM), or EPO906A (100, 1 and 0.01 nM) starting on day 1. Due to cell cycle kinetics, DNA synthesis (utilizing 3H-thymidine) was assessed on day 6. The degree of inhibition of 3H-thymidine incorporation at the high dose exposure for each agent was used to determine drug resistance, which was defined as <20% inhibition. Results and conclusions: Differential patterns of EDR were seen in ovarian and breast tumor samples for paclitaxel, XAA296A, and EPO906A. The degree of inhibition afforded by all three compounds was lower in the breast cancer samples compared to the ovarian cancer samples (paclitaxel 22% vs. 36% p = 0.03; XAA296A 21% vs. 35%, p = 0.06; EPO906A 21% vs. 36%, p = 0.03). A number of tumor samples were more resistant to paclitaxel compared to XAA296A and EPO906A, and slightly less than half of the paclitaxel-resistant specimens lacked cross-resistance to XAA296A or EPO906A. The resistance patterns for XAA296A in the breast and ovarian cancer samples appeared distinct compared to EPO906A. In conclusion, it is possible to demonstrate differential patterns of resistance to a variety of microtubule stabilizing agents in breast and ovarian cancer tumors samples. Further work will focus on identification of mechanisms of differential resistance to these agents.
Copyright © 2003 American Association for Cancer Research. Published as a Supplement to Clinical Cancer Research, Volume 9, Issue 16 (December 1, 2003).
A263 Combinations of the novel epothilone BMS-247550 with selected chemotherapeutic agents produce synergistic anti-neoplastic efficacy in preclinical in vitro and in vivo human cancer models.
Francis Y. Lee, Amy Camuso, Stephen Castaneda, Krista Fager, Ivan Inigo, David Kan, Kelly McGlinchey, and Mei-Li Wen. Bristol-Myers Squibb Co., Princeton, NJ.
The novel epothilone BMS-247550 is a potent microtubule stabilizing agent that has demonstrated superior antitumor activity than paclitaxel in a variety of preclinical human cancer models, including both paclitaxel sensitive and resistant variants. BMS-247550 is currently in Phase II clinical trials where it is being evaluated in a broad spectrum of cancer types. Significant activity had been observed in both taxane naive and resistant patient populations with advanced diseases. Based on these encouraging results a phase III program is planned where it is anticipated that BMS-247550 will be combined with existing chemotherapeutic agents. We conducted preclinical in vitro and in vivo studies to evaluate the combined effects of BMS-247550 with a variety of approved and experimental chemotherapeutic agents including cisplatin, capecitabine, a cdk2 inhibitor BMS-250636 and a farnesyltransferase (FT) inhibitor BMS-214662. Colony-formation assays were used to examine the cytotoxicity of BMS-247550 in combination with these agents in vitro. Using the human colon carcinoma cell line HCT116, isobologram analyses showed that the mode of interaction between BMS-247550 and other agents in vitro can be drug-, sequence- and dose-dependent, and can vary from synergism to antagonism. For BMS-214662, the ras FT inhibitor currently in Phase I clinical study, synergism was observed when BMS-247550 was administered first followed by BMS-214662. When cells were treated with BMS-247550 and BMS-214662 simultaneously or when BMS-214662 preceded BMS-247550, only additivity was observed. With BMS-250636, a CDK inhibitor, synergy was again observed when BMS-247550 was administered first, but antagonism was observed for the other two treatment sequences. For paclitaxel, all three treatment sequences resulted in additivity. In the case of cisplatin, additivity was observed when the two agents were used sequentially, but synergism was obtained for simultaneous treatment. In vivo studies were conducted in human tumor xenografts in nude mice. In the GEO human colon carcinoma model, BMS-247550 (iv, q4dx3) produced therapeutic synergism with capecitabine (po, qdx10) - the antitumor efficacy of the combination was superior than either of the agent administered alone at their respective maximum tolerated dose (MTD). With the FTI BMS-214662 in the HCT116/VM46 human colon carcinoma model, synergism was obtained when BMS-247550 treatment (iv, q4dx3) precedes the FTI (po, qdx10). Administering the two agents simultaneously resulted in additive effects. Taken together these data support the inclusion of BMS-247550 in clinical trials that utilize these chemotherapeutic agents in the treatment of cancer.
Copyright © 2003 American Association for Cancer Research. Published as a Supplement to Clinical Cancer Research, Volume 9, Issue 16 (December 1, 2003).
A270 Integrative toxicological and pharmacogenomic investigations of intestinal effects induced by epothilone b after single intravenous administration in rats.
Muriel Saulnier, Danielle Roman, Andreas Mahl, Jean-Claude Schaffner, Peter Ulrich, Paul M. J. McSheehy, Frank Staedtler, Markus Wartmann, Huseyin Firat, and Lutz Mueller. Novartis Pharmaceuticals Corp., East Hanover, NJ and Novartis Pharma AG, Basel, Switzerland.
EPO906 (epothilone B) is a novel natural product with potent cytotoxic activity foreseen for the treatment of solid tumors in humans. Diarrhea represents the most common adverse and dose-limiting effect with this compound in patients. EPO906 was intravenously administered as single bolus injection at 0.4 or 1.75 mg/kg to male Lewis rats to investigate the time- and dose-relationship between diarrhea, histopathological alterations in the intestinal tract and specific markers for intestinal tissue responses. Groups of 5 rats were sacrificed after a period of 4, 24 and 48 hours or 4 and 7 days. Investigations included clinical observations, macroscopic and microscopic examinations, gene expression analysis (cecum), immunofluorescence investigations of selected cytokines and COX?2 (jejunum and cecum), and quantitative determination of TNF-a (cecum).
Diarrhea, limited to the animals treated at 1.75 mg/kg, was present 4 to 7 days after administration. Histopathological examination of the intestinal tract revealed increased mitotic figures and apoptosis/single cell necrosis from 4 hours at 0.4 and 1.75 mg/kg. At 1.75 mg/kg, diffuse inflammation and distorted architecture were observed from 24 or 48 hours after administration. Dose-dependent increases in the number of cells expressing TNF-a and IFN-? were present in the jejunum from 4 or 24 hours. At 1.75 mg/kg, an increase in cells expressing COX-2 was noted in the cecum from 24 hours to 7 days. Quantitative measurements of TNF-a in the cecum revealed increased concentrations in single animals at 1.75 mg/kg after 4 and 7 days. Gene expression analysis in the cecum revealed both dose- and time-dependent changes in the mRNA expression levels reflecting mainly the pharmacological action of EPO906 on tubulin polymerization, stabilization of microtubules, cell cycle arrest, and induction of apoptosis. Changes in genes belonging to the acute phase protein family, coagulation cascade, tissue remodeling, angiogenesis, and lipid metabolism were indicative of an inflammatory response. Mainly at the high dose, there was a pronounced, early increase in the expression of these inflammatory genes, which remained high over time, contrary to the low dose. Major gene expression changes indicating a potential relationship to the manifestation of diarrhea included proinflammatory molecules (phospholipase a2 and the complement component c1q), cytokines and their receptors (TNF-r and IL-1 alpha), neuromediators (alpha2d, alpha2a adrenergic receptors and tachykinin 2), and some transporters like aquaporine 8.
In conclusion, the single intravenous administration of EPO906 to male lewis rats at 0.4 or 1.75 mg/kg caused dose- and time-dependent effects in the intestinal tract, which indicate a relationship with the expected antiproliferative activity of the compound. The dose of 0.4 mg/kg was well tolerated, whereas at 1.75 mg/kg intestinal lesions and inflammation led to diarrhea, possibly as a consequence of an impaired water/electrolyte and protective barrier function of the intestinal mucosa.
Copyright © 2003 American Association for Cancer Research. Published as a Supplement to Clinical Cancer Research, Volume 9, Issue 16 (December 1, 2003).
A276 Second generation epothilones and their biological properties.
Wolfgang Richter, Michael W. Cappi, Bernd Henkel, Sibylle Hess, Jean-Philippe Surivet, Daniel Baeschlin, Christian Hubschwerlen, Gerhard Höfle, Ludger Wessjohann, and Uwe Eichelberger. Morphochem AG, Munich, Germany, Morphochem AG, Basle, Switzerland, GBF(German Research Centre for Biotechnology), Braunschweig, Germany, and Institute of Plant Biochemistry, Halle, Germany.
Epothilones A and B are naturally occurring microtubule depolymerisation inhibitors, which inhibit the growth of human cancer cells in vitro at nano or even subnanomolar concentrations. Their attractive preclinical profile has made epothilones important lead structures in the search for improved cytotoxic anticancer drugs and hundreds of analogues and derivatives have been prepared and biologically characterised since their first description in a patent by the groups of Reichenbach and Höfle in 1993. Currently five compounds, natural epothilone B (EPO906 from Novartis), epothilone D or deoxyepothilone B (KOS-862 from Kosan Biosciences), the lactam analogue of epothilone B (BMS247550 from Bristol-Myers Squibb), the 21-substituted analogue of epothilone B (BMS-310705, also from Bristol-Myers Squibb) are undergoing clinical trials; another derivative from Schering AG finished pre-clinical development and entered phase I trials recently.
There are a number of synthetic and fermentation routes to the epothilones, however, Morphochem, in association with an academic institution, has identified a novel, convergent route to existing as well as to novel epothilones which has major advantages over the current routes. For the synthesis of the second generation epothilones several strategies are available and are suitable for obtaining the desired compounds in sufficient quantities required for pre-clinical and clincal development.
Clinical trials of the epothilone derivatives which are currently underway are assessing issues such as toxicity, efficacy and optimal dosage. The results from the pre-clinical and clinical studies published so far demonstrate that this compound class exhibits an enormous potential as anti-tumor drugs; nevertheless it seems quite obvious that the optimal representative of this class has not yet been found.
The greatest challenges, in our view are improvements and optimization work which needs to be made of the pharmaco-kinetic and the physico-chemical properties as well as securing enough supply.
Therefore, new analogs are presented which were designed by an rational approach and show improved properties. A new, extremely rapid synthesis of the "northern half" from the readily available starting materials acetoacetate and nerol was established. The enantioselective synthesis of the building blocks is performed by enzymatic resolution and asymmetric catalysis.
Their potential as novel second generation epothilones for the treatment of solid tumors is discussed.
Copyright © 2003 American Association for Cancer Research. Published as a Supplement to Clinical Cancer Research, Volume 9, Issue 16 (December 1, 2003).
C30 Complete ablation of xenograft tumors by a new class of epothilones: 9,10-dehydro-12,13-desoxyepothilones (dhdepos) and their derivatives.
Ting-Chou Chou, Huajin Dong, Alexey Rivkin, Fumihiko Yoshimura, Ana E. Gabarda, and Samuel J. Danishefsky. Memorial Sloan-Kettering Cancer Center, New York, NY.
Epothilones are potent microtubule-forming stabilizers and showed high anti-tumor efficacy in vitro and in vivo in nude mice-xenograft models. (Chou et al, Proc Natl Acad Sci USA 95:9642, 1998; 95:15798, 1998 & 98:8113, 2001). Several epothilones are currently under phase I/II clinical trials. We hereby report a new class of potent dHdEpos and their derivatives. 9,10-dehydro-12,13-desoxyepothilone B (dHdEpoB), 26-F3-dHdEpoB, deHEpoB and dHdEpoF inhibited human lymphoblastic CCRF-CEM leukemic cell growth with IC50 of 0.9 nM, 3.2 nM, 0.2nM and 0.5 nM, respectively, which compare favorably with dEpoB(3.6 nM), dEpoF(1.5 nM) and EpoB(0.5 nM). deHEpoB is now the most potent epothilone known in vitro and in vivo but showed relatively narrow therapeutic index. The optimal antitumor dose for dEpoB that achieved curative effect against MX-1 xenograft in nude mice was 30mg/kg, Q2Dx5, 6hr-i.v.infusion, however, for dHdEpoB the optimal dose for the same regimen was 3mg/kg. Formation of microtubule from tubulin at 37°C and 4°C were similar in the presence of 10µM of dHdEpoB and taxol, respectively. In addition, these formation were higher than the presence of GTP(1mM) alone at 4°C. 26-F3-9,10-dehydro-12,13-dEpoB achieves tumor disappearance in human xenografts in nude mice at relatively non-toxic or maximally tolerant doses (MTD) and total complete remission for over two months with early rapid recovery of body weight to the pretreatment control levels. Tumor as big as 2.3% of total body weight can be completely ablated.The therapeutic safety margin of F3-deH-dEpoB is one of the widest reported to date for a prospective cancer therapeutic agent in which 1/3 of MTD achieved total tumor remission. It is shown to be superior to any of the currently registered cancer therapeutic drugs evaluated in the same setting. The dramatic improvement in performance arises from increased cytotoxicity and efficacy associated with the E-9,10 double bond and superior metabolic stability provided by the three fluorine atoms at carbon 26. |
