Abstract 1074: Cellular uptake and metabolism of MT1-MMP-activated taxane prodrugs

ICT3205 is a targeted cytotoxic agent, comprised of an MT1-MMP-specific peptide and a paclitaxel warhead. Paclitaxel, originally approved in 1992, is one of the most commonly used chemotherapies for various cancer types, but is often characterized by its dose-limiting toxicity and low aqueous solubility. Tumor-targeted prodrugs are a promising strategy for the selective delivery of taxanes to prostate tumors, especially given the results of STAMPEDE trial1. Previous work has shown that ICT3205 is selectively metabolized in tumor tissue over the normal tissue, and that this is specifically mediated by MT1-MMP2. In vivo studies have demonstrated that when administered at an equimolar dose to paclitaxel, ICT3205 has a superior anticancer effect, with an absence of systemic toxicity. In this study we aim to further investigate the cellular uptake and metabolism of ICT3205 and other taxane analogues in MT1-MMP +ve (HT1080) and MT1-MMP -ve (MCF-7) cells in vitro. The compound will be also studied in the PC3 prostate cancer cell line, since it has been demonstrated before that there is a good correlation between MT1-MMP protein and mRNA expression. ICT3205 and analogues have been synthesized, and purified by semi-preparative HPLC. We report the tumor-specific metabolism of ICT3205 in prostate tumors ex vivo, and we are currently studying its cellular uptake and metabolism in MT1-MMP +ve and MT1-MMP -ve cancer cell-lines using HPLC, mass spectrometry and immunofluorescent microscopy. Moreover, ex vivo metabolism studies are performed on the analogues, while they will be also assessed for their metabolism and cellular uptake. These data will provide a better understanding of the mechanism of uptake and metabolism of this type of peptide prodrugs, and will provide the basis for the future synthesis and optimization of these prodrugs.

1.James ND et al. Addition of docetaxel, zoledronic acid, or both to first-line long-term hormone therapy in prostate cancer (STAMPEDE): survival results from an adaptive, multiarm, multistage, platform randomized controlled trial. Lancet. 2016;387(10024):1163-772.Loadman PM et al. Improved delivery of paclitaxel to prostate tumors: a membrane-type matrix metalloproteinase (MT-MMP)-targeted approach. Cancer Res. 2016;76:3

Citation Format: Athina Polykandritou, Alexandra Serre, Hannah Spencer, Goreti R. Morais, Amanda Race, Steven Shnyder, Paul Loadman, Robert A. Falconer. Cellular uptake and metabolism of MT1-MMP-activated taxane prodrugs [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1074.

Implementing subtype-specific pre-clinical models of breast cancer to study pre-treatment aspirin effects

Backgorund

Prior data suggest pre‐diagnostic aspirin use impacts breast tumour biology and patient outcome. Here, we employed faithful surgical resection models of HER2+ and triple‐negative breast cancer (TNBC), to study outcome and response mechanisms across breast cancer subtypes.

Method

NOD/SCID mice were implanted with HER2+ MDA‐MB‐231/LN/2‐4/H2N, trastuzumab‐resistant HER2+ HCC1954 or a TNBC patient‐derived xenograft (PDX). A daily low‐dose aspirin regimen commenced until primary tumours reached ~250 mm³ and subsequently resected. MDA‐MB‐231/LN/2‐4/H2N mice were monitored for metastasis utilising imaging. To interrogate the survival benefit of pre‐treatment aspirin, 3 weeks post‐resection, HCC1954/TNBC animals received standard‐of‐care (SOC) chemotherapy for 6 weeks. Primary tumour response to aspirin was interrogated using immunohistochemistry.

Results

Aspirin delayed time to metastasis in MDA‐MB‐231/LN/2‐4/H2N xenografts and decreased growth of HER2⁺/TNBC primary tumours. Lymphangiogenic factors and lymph vessels number were decreased in HER2⁺ tumours. However, no survival benefit was seen in aspirin pre‐treated animals (HCC1954/TNBC) that further received adjuvant SOC, compared with animals treated with SOC alone. In an effort to study mechanisms responsible for the observed reduction in lymphangiogenesis in HER2⁺ BC we utilised an in vitro co‐culture system of HCC1954 tumour cells and mesenchymal stromal cells (MSC). Aspirin abrogated the secretion of VEGF‐C in MSCs and also decreased the lymph/angiogenic potential of the MSCs and HCC1954 by tubule formation assay. Furthermore, aspirin decreased the secretion of uPA in HCC1954 cells potentially diminishing its metastatic capability.

Conclusion

Our data employing clinically relevant models demonstrate that aspirin alters breast tumour biology. However, aspirin may not represent a robust chemo‐preventative agent in the HER2⁺ or TNBC setting.

Abstract 4926: Preclinical intratumoral pharmacokinetics (PK) of capecitabine (Cap) given +/- eribulin (Eri)

Background: Eri & Cap are active as single agents in metastatic breast cancer, but the combination appears particularly active & well tolerated clinically. We explored the hypothesis that normalisation of tumor vasculature by Eri seen in preclinical models may enhance delivery of Cap to the tumor when given in combination.

Methods: Mice bearing MDA-MD-231 xenografts were treated with Cap 540mg/kg p.o. day 1 alone or following Eri 1mg/kg i.v. on day -8 then sacrificed on day 1 at 15min, 30min, 1h, 2h, 4h, 6h or 24h after Cap (n=3/ time point). Concentrations of Cap & its metabolites (5'-deoxy-5-fluorocytidine, 5DFCR; 5-Fluoro-5'-deoxyuridine, 5DFUR; 5-Fluorouracil, 5FU; 5-Fluorouridine, 5FUrd; & 5-Fluoro-2'-deoxyuridine, 5FdUrd) in plasma, tumor, skin & muscle were analysed by LC-MS/MS. Half the tumor was reserved to study Cap distribution using matrix-assisted laser desorption/ionization mass spectrometry imaging with haem imaged as a surrogate marker for tumour vasculature.

Results: We found no differences in the plasma, skin, muscle or tumor PK profiles of Cape & its metabolites with the addition of Eri (Table 1). In both treatment arms intratumoral concentrations of 5DFUR, 5FU, 5FUrd & 5FdUrd appeared higher in tumors while those of Cape & 5DFCR appeared higher in normal healthy tissues. To mitigate the variability in tumor Cape concentrations between animals, we also expressed the results as the ratio of drug concentration in tumour:healthy tissue but again found no apparent effect of Eri on intratumoral drug concentrations. Analyses of drug distribution are on-going & will be presented.

Conclusion: No clear effect of Eri on intratumoral concentrations of Cape or its metabolites was seen. We are exploring potential differences in tumour drug distribution.

Citation Format: Maria Jove Casulleras, Jade Spencer, Paul Loadman, Malcolm Clench, Steve Shnyder, Patricia Cooper, Cristina Russo, Amanda Race, Ramon Salazar, Chris Twelves. Preclinical intratumoral pharmacokinetics (PK) of capecitabine (Cap) given +/- eribulin (Eri) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4926.

Structural properties governing drug-plasma protein binding determined by high-performance liquid chromatography method

The high-performance liquid chromatography (HPLC) method employing stationary phases immobilized with plasma proteins was used for this study to investigate the structural properties governing drug-plasma protein binding. A set of 65 compounds with a broad range of structural diversity (in terms of volume, hydrogen-bonding, polarity and electrostatic force) were selected for this purpose. The Abraham linear free energy relationship (LFER) analyses of the retention factors on the immobilized HSA (human serum albumin) and AGP (α₁-acid glycoprotein) stationary phases showed that McGowan's characteristic molecular volume (V), dipolarity/polarizability (S) and hydrogen bond basicity (B) are the three significant molecular descriptors of solutes determining the interaction with immobilized plasma proteins, whereas excess molar refraction (E) is less important and hydrogen bond acidity (A) is not of statistical significance in both systems, for electrically neutral compounds. It was shown that ionised acids, as carboxylate anions, bind very strongly to the immobilized HSA stationary phase and that ionised bases, as cations bind strongly to the AGP stationary phase. This is the first time that the effect of ionised species on plasma protein binding has been determined quantitatively; the increased binding of acids to HSA is due almost entirely to acids in their ionised form.

Abstract 4205: Intracellular pharmacokinetics of 5FU and palbociclib: Uptake and efflux in disaggregated cells and 3D models

Introduction:

Pre-clinical drug development does not routinely assess drug penetration beyond use of the human intestinal Caco-2 cell line as an in vitro model of the intestinal barrier to evaluate potential drug absorption. Detailed in vivo tumour pharmacokinetic (PK) studies are unusual. However, the ability of a drug to reach its target plays a key role in drug efficacy.

Here we present a methodology to study drug PK in disaggregated cells and in a 3D cell culture model (spheroids) using 5FU and palbociclib. The comparison of these 2 drugs is of interest taking into account their different mechanisms and chemical properties, logP 0.58 and 2.12 respectively. Knowledge of cellular uptake and efflux in disaggregated cells and 3D models together with intracellular binding characteristics is vital to optimise a better drug delivery.

Experimental procedures:

MCF-7 and DLD-1 cells were used for cellular and spheroid experiments. Spheroid growth was optimized to produce a spheroid of 300-400µm diameter at day 4. Drug uptake: 1x106 cells or 30 spheroids/time point were exposed to 10µM 5FU or 0.1µM of palbociclib for different time durations. For 5FU, drug uptake at 100µM was also monitored to detect 5FU metabolites. Drug efflux: 1x106 cells were exposed first to 100µM 5FU or 0.1µM of palbociclib for 60min then, drug media removed, substituted by fresh media and sample over a 4h period. Drug concentration inside cells and in the fresh media was measured. Drug concentrations were measured using HPLC-Tandem Mass Spectrometry. Each experiment was carried out in triplicate.

Results:

5FU achieves intracellular steady state within 5-10min in cells and spheroids with equal concentrations inside and outside the cell/spheroids. Intracellular metabolites of 5FU were only seen with 100µM treatment and had a similar PK characteristics to 5FU. 5FU drug efflux experiments showed that efflux occurred rapidly within 5min achieving a new equilibrium with 30% of 5FU remaining inside the cells. 5FUrd was also detected extracellularly after 1h in the fresh media, showing that cells have both 5FU and metabolite efflux. In contrast, palbociclib cellular uptake had an initial intracellular drug concentration peak at 5-10min which then plateaued. Interestingly, total intracellular concentrations of palbociclib were over 30 times higher compared to the external media drug at steady state. Spheroid and efflux experiments of palbociclib are on-going and will be presented.

Conclusions:

Our results suggest that both drugs penetrate quickly in cancer cells. Palbociclib intracellular concentrations were 30 times higher than the external media suggesting excessive intracellular binding while 5FU achieve equal concentrations between the inside and outside of the cell. These data and methodology may be useful to allow generation of mathematical models to improve drug delivery design.

Citation Format: Maria Jove, Paul Loadman, Jade Spencer, Lava Sulayman, Jo Wicks, Amanda Race, Chris Twelves. Intracellular pharmacokinetics of 5FU and palbociclib: Uptake and efflux in disaggregated cells and 3D models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 4205. doi:10.1158/1538-7445.AM2017-4205

Abstract 1576: Analysis of HSP10 as a putative biomarker of breast cancer

Introduction. Detection of the very earliest stages of breast cancer remain challenging and the differentiation of those carcinomas that develop into aggressive invasive forms from those that remain benign has not been achieved. Consequently, removal of the tumor by total mastectomy, local excision (LE) plus adjuvant radiotherapy (RT), or LE alone remain the main treatments, even if they prove unnecessary. Previously, using proteomics analysis, we identified proteins that were significantly changed in expression between matched normal and tumor tissues from Cypriot patients with different stages of the disease. Heat shock protein 10 (HSP10), which is responsible for the mitochondrial protein folding, was one of the proteins identified. A range of methods were used to verify HSP10 as a possible biomarker for breast cancer.

Materials and Methods. Protein extracts from breast cancer cell lines and breast tissue biopsies from Cypriot patients with different stages of the disease were analysed by (i) Western Blotting (WB) and (ii) multiple reaction monitoring mass spectrometry (MRM MS) using two proteotypic peptides unique to HSP10, in proteins extracted form fresh frozen as well as from FFPE breast tissues. Tissue microarray analysis (TMA) was carried out on core biopsies from the Leeds Breast Tissue Bank.

Results. WB indicated that HSP10 was ubiquitously and evenly expressed across breast cancer cell lines representative of different phenotypes. WB and MRM MS revealed that expression was significantly increased in invasive carcinoma compared to normal tissue, but was patient specific for fibroadenoma and DCIS cases. TMA of 360 samples indicated a strong correlation between staining and type of breast cancer (invasive ductal and lobular), grade and ER positive phenotype.

Discussion. Proteomics analysis using iTRAQ quantitation indicated an average of 1.6 fold increase in HSP10 in invasive carcinomas compared to matched normal tissue. By comparison, WB and MRM MS analysis indicated a significantly amplified discrimination in the expression of HSP10 as has been observed with other biomarker candidates. HSP10 protein levels were in agreement with mRNA results observed in breast cancer cell lines reported in the Cancer Cell Line Encyclopaedia. In conclusion, HSP10 represents an excellent biomarker candidate for further validation.

Citation Format: Sadr-ul Shaheed, Andreas Hadjisavvas, Kleitos Sokratous, Paul Loadman, Chris Sutton, Kyriacos Kyriacou. Analysis of HSP10 as a putative biomarker of breast cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1576. doi:10.1158/1538-7445.AM2015-1576

Identification of stage-specific breast markers using quantitative proteomics

Matched healthy and diseased tissues from breast cancer patients were analyzed by quantitative proteomics. By comparing proteomic profiles of fibroadenoma (benign tumors, three patients), DCIS (noninvasive cancer, three patients), and invasive ductal carcinoma (four patients), we identified protein alterations that correlated with breast cancer progression. Three 8-plex iTRAQ experiments generated an average of 826 protein identifications, of which 402 were common. After excluding those originating from blood, 59 proteins were significantly changed in tumor compared with normal tissues, with the majority associated with invasive carcinomas. Bioinformatics analysis identified relationships between proteins in this subset including roles in redox regulation, lipid transport, protein folding, and proteasomal degradation, with a substantial number increased in expression due to Myc oncogene activation. Three target proteins, cofilin-1 and p23 (increased in invasive carcinoma) and membrane copper amine oxidase 3 (decreased in invasive carcinoma), were subjected to further validation. All three were observed in phenotype-specific breast cancer cell lines, normal (nontransformed) breast cell lines, and primary breast epithelial cells by Western blotting, but only cofilin-1 and p23 were detected by multiple reaction monitoring mass spectrometry analysis. All three proteins were detected by both analytical approaches in matched tissue biopsies emulating the response observed with proteomics analysis. Tissue microarray analysis (361 patients) indicated cofilin-1 staining positively correlating with tumor grade and p23 staining with ER positive status; both therefore merit further investigation as potential biomarkers.

A phase I pharmacokinetic (PK) and pharmacodynamic (PD) study of the selective aurora kinase inhibitor GSK1070916A

2525

Background: GSK1070916A is a potent and selective inhibitor of Aurora B and C. This phase I study in collaboration with GlaxoSmithKline was part of the Cancer Research UK Clinical Development Programme. Methods: Patients (pts) with advanced/metastatic solid cancers for whom there was no standard therapy, with adequate performance status and organ function were eligible for GSK1070916A (1 hour i.v. infusion days 1 – 5, every 21 days). The primary objectives were to determine the safety profile, dose limiting toxicity (DLT) and maximum tolerated dose (MTD) of GSK1070916A. The starting dose was 5mg/m2/day, with initial single pt cohorts, followed by “3 + 3” cohorts and expansion at the MTD. DLTs included prolonged (> 5 days) or complicated grade 4 neutropenia; the MTD was the highest dose at which < 1 of 3 - 6 pts experienced DLT. Cycle 1 blood and healthy skin biopsies were obtained for PK and PD assays. The expanded cohort included 6 pts having pre- and post-treatment functional imaging studies (FDG PET-CT and MRI), and a further 6 having paired tumour biopsies for PD studies. Results: Nine single pt cohorts received up to 73mg/m2/day of GSK1070916A with no grade 3 or 4 related adverse events. At 102.2mg/m2/day, 1 pt had a DLT (febrile neutropenia) and 2 pts non-DLT grade 4 neutropenia; this dose was considered unacceptably toxic and 23 pts received a lower dose of 85mg/m2/day; 7/23 pts had prolonged/complicated grade 4 neutropenia, 5 of whom continued GSK1070916A with dose reduction +/- delay. There were no treatment related deaths. A pt with ovarian cancer (102.2mg/m2/day) had a RECIST PR; 19 pts had stable disease for < 223 days. GSK1070916A PK were linear with a strong correlation between exposure (AUC) and reduction in neutrophils (r2 0.91). At the 85 mg/m2 dose, mean day 1 t1/2 was 8.98 hours and Cl 9.2 l/h; AUCinf was 10% higher on day 5 than day 1. PD results in healthy skin (phosphoHistone-H3, Ki 67 and cleaved caspase-3) were inconsistent. Conclusions: The MTD of GSK1070916A as a 1 hour i.v. infusion on days 1 – 5, every 21 days is 85mg/m2/day, with predictable and manageable neutropenia as the DLT and evidence of clinical activity. Serum levels of cytokeratin-18, tumour PD and functional imaging data will be presented. Clinical trial information: NCT01118611.

Abstract 2505: Expression profiling of cofilin-1 in breast cancer cell lines and biopsies

Introduction. Despite oestrogen and human epidermal growth factor receptor (HER2) targeted therapies improving patient survival, breast cancer continues to be a significant cause of premature death since many tumors do not respond to treatment or acquire resistance. Hence there remains the need to identify novel targets for therapy and biomarkers for prediction of response to treatment. A quantitative proteomics study of matched normal and tumor biopsies identified 63 proteins to be significantly increased or decreased in stage-specific tumors. Some have previously been associated with breast cancer, while others such as cofilin-1 represent new targets for investigation. Cofilin-1 was subject to a range of analyses to determine its association with breast cancer.

Materials and Methods. Western blotting (WB) was performed on protein extracts from breast cancer cell lines and matched normal and tumor biopsies from Cypriot patients with different stages of the disease. Immunohistochemistry (IHC) was carried out on core biopsies from the Leeds Breast Tissue Bank. Multiple reaction monitoring (MRM) mass spectrometry was performed on trypsin-digested protein extracts from biopsies. Data from protein (Human Protein Atlas) and genomics (BioGPS, TiGER, UniGene) databases were assimilated with the experimental results.

Results. WB indicated that cofilin-1 was ubiquitously expressed in tumor cell lines, representative of Luminal A, Luminal B, basal-like, claudin-low and HER2 phenotypes, at higher levels than normal breast cell lines. WB also indicated increased expression of cofilin-1 in invasive carcinoma tissues, compared to matched normal. Patients with ductal carcinoma in situ or fibroadenoma exhibited less clear results, either increasing slightly or remaining unchanged. MRM analysis of three cofilin-1 peptides in tissue extracts of invasive carcinoma patients indicated expression only in tumor. IHC of core biopsies exhibited strong staining for cofilin-1 in ductal invasive carcinoma tissues with no staining in normal breast cells. Genomics and proteomics databases indicate that cofilin-1 is expressed in a diverse range of normal tissues including major organs, gastro-intestinal tract, skin and tonsils. However, mRNA and protein levels are observed to be increased in skin, lung, liver, pancreatic and breast tumors.

Discussion. The original quantitative proteomics data indicated only relatively small changes in expression (less than 2-fold) due to dynamic range limitations of the technique. Additional analytical approaches substantiated that cofilin-1 is significantly up-regulated in advanced breast cancer patients. Further patient sets are required to confirm the importance of cofilin-1 levels in early stages of breast cancer compared to benign tissues. Bioinformatics provided further confidence in our findings, highlighting the value of utilising databases for evidence of disease-specific proteins.

Citation Format: Chris Sutton, Sadr ul-Shaheed, Paul Loadman, Valerie Speirs, Andrew Hanby, Andreas Hadjisavvas, Kyriacos Kyriacou. Expression profiling of cofilin-1 in breast cancer cell lines and biopsies. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2505. doi:10.1158/1538-7445.AM2013-2505

Evaluation of the safety of C-1311 administered in a phase 1 dose-escalation trial as a 1-hour infusion once every 3 weeks in patients with advanced solid tumors

12005

Background: C-1311 (Symadex) is the most active member of a new series of anti-cancer agents, the imidazoacridinones, specially designed compounds developed from research on the structure-activity relationships of existing cancer therapies. This clinical trial was designed to assess the safety profile of ascending doses of C-1311 and to determine the recommended dose when administered every 3 weeks (Q3W). Methods: Patients with advanced solid tumors refractory to conventional therapy were enrolled. The dose escalation scheme was divided into 2 phases with different schedules. Initially, C-1311 was given as a 1-hour daily infusion over 5 consecutive days. Following review of pre-clinical data, coupled with preliminary results of a parallel weekly study, a day 1 only schedule was explored. The maximum tolerated dose (MTD) is defined as the dose at which 2/3 or 2/6 pts experience a dose-limiting toxicity (DLT). The recommended dose (RD) is defined as the dose level below the MTD, confirmed by expansion of the RD cohort to 9 pts. Results: The first 5 patients were treated at doubling doses of 6, 12, 24, 48 and 96 mg/m2 on a schedule of 1-hour daily infusions over 5 consecutive days, repeated every 3 weeks; no DLTs were seen. The 6th pt received the same total dose as the 5th (480 mg/m2) but on day 1 only, following which a modified Fibonacci design with cohorts of 3 pts and 33% dose increments was utilized. Subsequently, pts received C-1311 at doses of 640 mg/m2 (3 pts), 850 mg/m2 (3 pts) and 1100 mg/m2 (5 pts); 2/5 pts treated at 1100 mg/m2 experienced an identical DLT (grade 4 neutropenia for ≥7 days with grade 4 thrombocytopenia), defining this as the MTD. The 850 mg/m2 cohort was expanded to confirm this as the RD. Neutropenia, thrombocytopenia, anemia and fatigue have been reported as drug-related grade 3 or 4 adverse events (AEs). Grade 1 or 2 AEs most commonly described as drug-related are fatigue, nausea, vomiting and diarrhea. Stable disease was observed in two patients with advanced malignancy. A pt had a stable disease over 8 cycles and one over 6 cycles. Plasma drug concentrations are linear and proportional to dose. Conclusions: A C-1311 dose of 850 mg/m2 administered once every 3 weeks offers a predictable and tolerable safety profile.

[Table: see text]

Evaluation of the safety of C-1311 administered in a phase I dose-escalation trial as a weekly infusion for 3 consecutive weeks in patients with advanced solid tumors

2069

Background: C-1311 is the most active member of a new series of anti-cancer agents, the imidazoacridinones, specially designed compounds developed from research on the structure-activity relationships of other cancer therapies such as the anthracenediones (e.g. mitoxantrone) and the anthracyclines (e.g. doxorubicin). This first-in-man clinical trial was designed to assess the safety profile of C-1311 and determine the recommended dose upon weekly administration for 3 consecutive weeks, within a 28-day cycle. Methods: Patients with advanced solid tumors refractory to conventional therapy were enrolled. The dose escalation scheme was divided into 2 phases: an initial accelerated phase of cohorts of 1 patient at doubling doses and a modified Fibonacci phase of cohorts of 3 pts at 25–33% dose increments. The maximum tolerated dose (MTD) was defined as the dose at which 2/3 or 2/6 pts experienced a dose-limiting toxicity (DLT). The recommended dose (RD) was defined as the dose level below the MTD, confirmed by expansion of the cohort to 9 pts. Results: 16 pts received doses of 15, 30, 60, 120, 240, 480 and 640 mg/m2/wk. The 2 pts treated at 640 mg/m2 experienced a DLT (grade 3 neutropenia preventing the administration of the 2nd and 3rd dose of the first cycle respectively). The RD was defined as 480 mg/m2/wk (cohort of 9 pts). Six additional pts were treated at the RD in an extension study in which one of the first 2 doses was given orally. Overall, 2 serious adverse events have been reported as possibly drug-related, both consisting of post-infusion fever without evidence of infection, resolving within 24 hours. Transient neutropenia was the only recurring grade 3 or 4 drug-related adverse event (AE). Grade 1 or 2 AEs most commonly reported as drug-related were nausea, asthenia, vomiting and diarrhea. Stable disease was observed in 3 patients with advanced malignancy. One pt had a stable course over > 8 cycles, and two had stable courses over 4 cycles. Plasma drug concentrations were linear and proportional to dose. Conclusions: The recommended phase 2 dose of 480 mg/m2, weekly × 3 q 28 days, offers a predictable safety profile and excellent tolerability.

[Table: see text]

5,6-dimethylxanthenone-4-acetic acid (DMXAA), a novel antivascular agent: phase I clinical and pharmacokinetic study

The purpose of this phase I, dose-escalation study was to determine the toxicity, maximum tolerated dose, pharmacokinetics, and pharmacodynamic end points of 5,6-dimethylxanthenone acetic acid (DMXAA). In all, 46 patients received a total of 247 infusions of DMXAA over 15 dose levels ranging from 6 to 4900 mg x m(-2). The maximum tolerated dose was established at 3700 mg x m(-2); dose-limiting toxicities in the form of urinary incontinence, visual disturbance, and anxiety were observed at the highest dose level (4900 mg x m(-2)). The pharmacokinetics of DMXAA were dose dependent. Peak concentrations and area under the curve level increased from 4.8 microM and 3.2 microM h, respectively, at 6 mg x m(-2) to 1290 microM and 7600 microM h at 3700 mg x m(-2), while clearance declined from 7.4 to 1.7 l h(-1) x m(-2) over the same dose range. The terminal half-life was 8.1+/-4.3 h. More than 99% of the drug was protein bound at doses up to 320 mg x m(-2); at higher doses the percent free drug increased to a maximum of 6.9% at 4900 mg x m(-2). Dose-dependent increases in the serotonin metabolite 5-hydroxyindoleacetic acid were observed at dose levels of 650 mg x m(-2) and above. There was one unconfirmed partial response at 1300 mg x m(-2). In conclusion, DMXAA is a novel vascular targeting agent and is well tolerated.