Phase I Clinical Trial of Weekly Combretastatin A4 Phosphate: Clinical and Pharmacokinetic Results
By Gordon J.S. Rustin, Susan M. Galbraith, Helen Anderson, Michael Stratford, Lisa K. Folkes, Luiza Sena, Lindsey Gumbrell, and Patricia M. Price
Purpose: A phase I trial was performed with combret- astatin A4 phosphate (CA4P), a novel tubulin-binding agent that has been shown to rapidly reduce blood flow in animal tumors.
Patients and Methods: The drug was delivered by a 10-minute weekly infusion for 3 weeks followed by a week gap, with intrapatient dose escalation. Dose escalation was accomplished by doubling until grade 2 toxicity was seen. The starting dose was 5 mg/m2.
Results: Thirty-four patients received 167 infusions. CA4P was rapidly converted to the active combretastatin A4 (CA4), which was further metabolized to the glucuronide. CA4 area under the curve (AUC) increased from 0.169 at 5 mg/m2 to 3.29 umol . h/Lat 114 mg/m2. The mean CA4 AUC in eight patients at 68 mg/m2 was 2.33 umol · h/L com- pared with 5.8 umol . h/L at 25 mg/kg (the lowest effective
dose) in the mouse. The only toxicity that possibly was related to the drug dose up to 40 mg/m2 was tumor pain. Dose-limiting toxicity was reversible ataxia at 114 mg/m2, vasovagal syncope and motor neuropathy at 88 mg/m2, and fatal ischemia in previously irradiated bowel at 52 mg/m2. Other drug-related grade 2 or higher toxicities seen in more than one patient were pain, lymphopenia, fatigue, anemia, diarrhea, hypertension, hypotension, vomiting, vi- sual disturbance, and dyspnea. One patient at 68 mg/m2 had improvement in liver metastases of adrenocortical car- cinoma.
Conclusion: CA4P was well tolerated in 14 of 16 patients at 52 or 68 mg/m2; these are doses at which tumor blood flow reduction has been recorded.
J Clin Oncol 21:2815-2822. @ 2003 by American Society of Clinical Oncology.
C OMBRETASTATIN A4 phosphate (CA4P; Cancer Re- search Campaign trial 96-04) [cis-1-(3,4,5,-trimethoxy- phenyl)-2-(4’-methoxyphenyl) ethene-3’-0-phosphate, disodium salt] is a more water-soluble prodrug of combretastatin A4 (CA4), which was isolated from the root bark of the South African bush willow (Combretum caffrum) after Pettit et al1-3 learned that extracts of this tree had been used by the Zulus as herbal remedies and as paint for spears. CA4P is converted into CA4 by nonspecific endogenous phosphatases present in plasma and on endothelial cells. It has a structure similar to that of colchicine, and binds to the colchicine binding site on tubulin.4 After studies on colchicine and the vinca alkaloids, which demonstrated vascular targeting activity despite their different binding sites on tubulin, additional experi- ments on novel tubulin-binding agents were performed.5
The initial in vivo studies demonstrated that CA4 reduced the tumor blood flow rate 24 hours after injection.5 These studies were confirmed by additional in vivo studies using CaNT tumors in mice, in which a single dose of 100 mg/kg caused 93% reduction in tumor vascular volume and extensive hemorrhagic necrosis.6 As with other vascular targeting agents, however, a rim of viable tumor cells remained, and these cells were sufficient to rapidly repopulate the tumor, so that a single dose produced minimal delay in tumor growth.7 However, twice-daily dosing did produce growth delays (S.A. Hill, personal commu- nication, July 2002). The combination of CA4P with cisplatin or radiation causes enhanced growth delay, and cell death with no recurrence was achieved when it was administered after a radiolabeled antibody to carcinoembryonic antigen in a colorec- tal xenograft model.7,8
In an ex vivo isolated tumor perfusion system, CA4P rapidly caused a more than three-fold increase in tumor vascular
resistance but no change in vascular resistance in a similarly isolated limb perfusion.6,9 In P22 carcinosarcomas in rats, a 100-fold decrease in tumor blood flow rate was seen at 6 hours, with a much smaller reduction to blood flow rate in spleen, skin, skeletal muscle, and brain. No significant reduction was ob- served in heart, kidney, and small intestine.9 In vitro studies have shown a concentration-dependent action, with much greater cytotoxic and antiproliferative activity against proliferating human endothelial cells compared with that against quiescent endothelial cells or human breast carcinoma cells.6 Disruption of endothelial cellular networks in a collagen layer was seen after treatment with CA4 or the phosphate, CA4P. 10 CA4P also produced a rapid increase in permeability of endothelial cell monolayers, which was enhanced in the presence of tumor-conditioned medium.11 This was associated with a disruption of the actin and tubulin cytoskeleton, and a marked
From the Department of Medical Oncology and Gray Cancer Institute, Mount Vernon Hospital, Northwood, Middlesex; Cancer Research United Kingdom Positron Emission Tomography Oncology Group, Hammersmith Hospital, London; and Cancer Research United Kingdom, London, United Kingdom. Submitted May 29, 2002; accepted March 7, 2003.
This study was performed on behalf of the phase I to phase II trials committee of Cancer Research United Kingdom and was supported by grants from the Cancer Research United Kingdom and The Community Fund.
Address reprint requests to Gordon Rustin, MD, Department of Medical Oncology, Mount Vernon Hospital, Northwood, Middlesex HA6 2RN, United Kingdom; email: rustin@mtvern.co.uk.
@ 2003 by American Society of Clinical Oncology. 0732-183X/03/2115-2815/$20.00
change in endothelial cell shape,12 which occurred in a similar rapid time scale to that of the in vivo blood flow changes.
Unlike colchicine and the other tubulin-binding agents, the doses at which vascular targeting activity was observed were well below the maximum-tolerated dose (MTD) in mice. This was therefore the first tubulin-binding agent that appeared to have a wide enough therapeutic window to achieve tumor vascular targeting effects at tolerable doses. Phase I trials were planned to investigate three different schedules after toxicity studies of single-dose, daily × 5, and weekly × 4 regimens were performed using mice, rats, and dogs by OXIGENE, Inc (Wa- tertown, MA).13,14 The organ systems affected by CA4P were within the gastrointestinal tract, with only minor effects on liver, bone marrow, or renal function. Transient weakness in the hind limbs and bradycardia was noted in the dogs. The estimated single-dose MTD was 360 mg/m2 in rats but 100 mg/m2 in dogs. On a weekly × 4 schedule the estimated MTD in rats was 300 mg/m2. The recommended starting dose in humans (one third of the lowest toxic dose in dogs) was 5 mg/m2.
The United Kingdom trial using a weekly schedule of CA4P was the first trial of a vascular targeting agent in which measurement of tumor blood flow parameters by either positron emission tomography (PET) or dynamic contrast enhanced magnetic resonance imaging (DCE-MRI) scans was planned for all patients from the outset. The results of these surrogate markers of drug activity are published elsewhere.15,16 The dose-escalation schedule followed an accelerated titration design with the aim of treating fewer patients at doses below the biologically active level while maintaining adequate safety precautions.17 The primary objective was to determine the toxicity profile of CA4P; secondary objectives were to determine the pharmacokinetics and effects of CA4P on tumor blood flow and possible antitumor effects. On the basis of tumor blood flow effects and the MTD, we recommend a dose for additional evaluation in humans.
PATIENTS AND METHODS
Eligibility
Patients with histologically confirmed cancer who were not amenable to standard curative therapy or were refractory to conventional therapy were eligible for this study. In addition, patients had to have a tumor suitable for PET or MRI examination. This required a mass ≥ 2 cm in diameter; for MRI examination, the mass had to be in a position that did not move with respiration. Other eligibility requirements included World Health Organiza- tion performance status of 0 to 2; life expectancy ≥ 4 months; age ≥ 18 years; adequate hepatic, renal, and hematologic function; absence of other anticancer therapy for 4 weeks; and no active concurrent malignancies except cone-biopsied carcinoma-in-situ of cervix or adequately treated basal or squamous carcinoma of skin. Patients were excluded if they had received previous irradiation to the tumor to be imaged by MRI or PET, or previous irradiation to they tumor mass to be assessed for response or by functional imaging. An amendment excluded patients who had received any radiother- apy after a toxic event in the final expansion of the 52 mg/m2 dose level. Patients were also excluded if they were taking heparin, warfarin, or the nonsteroidal anti-inflammatory agent naproxen (which interferes with the pharmacokinetic assay); were pregnant, lactating, or at risk of pregnancy; had active brain metastasis, diabetes, or other serious medical condition; had autoimmune disorders, inflammatory bowel disease, or ischemic heart
disease; or had open surgery or serious infection in the previous 28 days. All patients gave written informed consent.
Study Design, Dosage, and Drug Administration
CA4P was supplied as a freeze-dried powder containing 100 mg of CA4P. Because CA4P is moderately sensitive to temperature and light, the vials were stored at 2 to 8℃ in the dark and removed from cold storage only before use. The lyophilized product was aseptically reconstituted with 2 mL of 0.9% sodium chloride for injection to provide an isotonic solution containing 50 mg/mL of CA4P. The reconstituted solution had to be clear and free of particles. The reconstituted product and any solutions prepared from it were required to be used within 24 hours from the time of initial reconstitution and stored at 4℃ in the dark. The solution could be mixed with 0.9% NaCl for administration to a volume of approximately 100 mL (minimum volume 100 mL; maximum volume 150 mL). CA4P was supplied to Cancer Research United Kingdom (CRUK) by OXIGENE and distributed by Gavin Halbert, PhD, CRUK Formulation Unit, University of Strathclyde, Glasgow, United Kingdom.
Approval was obtained for this study from the research ethics committees of Mount Vernon Hospital (Middlesex, United Kingdom) and Hammersmith Hospital (London, United Kingdom). The starting dose was 5 mg/m2. The drug was administered using a volumetric pump over a 10-minute period. A cycle of treatment consisted of one infusion of CA4P each week for 3 weeks with a 1-week gap before the next cycle. Once intrapatient dose escalation ceased, patients could receive additional cycles of treatment if they had had a significant reduction in tumor blood flow as assessed by MRI or PET imaging, stable or responding disease, and grade 2 or lower drug-related toxicity that recovered within 2 weeks. Toxicities were recorded according to the National Cancer Institute (NCI) common toxicity criteria version 2.
The dose-escalation schedule followed an accelerated titration design. Dose escalation began by doubling the dose until one grade 2 or higher drug-related toxicity (except for alopecia and nausea or vomiting) occurred, when subsequent escalation would be no greater than 1.3 times the present dose. Intrapatient dose escalation occurred; the plan was for each patient to receive one cycle (three infusions) at a maximum of three dose levels. An individual patient was only allowed to receive an escalated dose if they experienced no drug-related toxicity of grade 2 or higher. New and current patients were allowed to receive an escalated dose after one cycle provided there had been fewer than two instances of grade 2 or higher drug-related toxicity (except for alopecia and nausea or vomiting). Treatment of an additional three patients was planned at doses at which blood flow reductions were seen but no toxicity of grade 2 or higher occurred. If one patient experienced a dose-limiting toxicity (DLT), up to three additional patients were to be recruited at that dose level. Once two patients had experienced a DLT, at least six patients were to be treated at a lower dose to determine the MTD.
DLT was defined as drug-related toxicity with grade 2 or higher neuro- toxicity, grade 3 or 4 other nonhematologic toxicity (excluding alopecia, nausea, vomiting, diarrhea, and tumor pain in patients who had not received optimal treatment with antiemetics, antidiarrheals, or analgesics), or grade 4 hematologic toxicity lasting more than 4 days or associated with grade 2 or more fever, infection, or bleeding. The MTD was defined as a dose below that at which 30% or more of the patient population (at least two of six patients) would suffer DLT as a result of the drug. The trial allowed additional expansion of cohorts below the MTD to assess blood flow effects. Blood flow measurements were planned to be performed before and after the first infusion of each cycle. At each dose level at least one patient was assessed by DCE-MRI and one patient was assessed by PET.
Criteria for evaluating safety included the recording of signs and symp- toms at baseline and during the study period, and laboratory tests including hematology, biochemistry, and urinalysis. Blood pressure and pulse were monitored every 30 minutes for the first 4 hours. ECGs were performed before treatment, hourly for the first 8 hours after the first dose, and at the first dose of each new dose level. The last five patients in the trial also had a 24-hour ECG tape recorded on the first dose because of reports of QTc prolongation in another trial of CA4P. Any adverse event that occurred while a patient was enrolled onto the trial was recorded and graded using the National Cancer Institute common toxicity criteria version 2.0. The relation-
PHASE I CLINICAL TRIAL OF CA4P
ship of the adverse event to the drug treatment was scored as almost certainly, probably, possibly, unlikely, or unrelated. All assessable sites of disease were measured before treatment, before each subsequent cycle of three infusions, and at the end of the study. Patients having DCE-MRI scans also had tumor measurements before each cycle of three infusions. Response was assessed by criteria that were based on those produced on the initiative of the World Health Organization.
Pharmacokinetics
Blood samples in 5-mL EDTA-coated tubes were taken before treatment; 1 minute before the end of infusion; every 15 minutes for the first hour; then at 1.5, 2, 4, 8, 12, and 24 hours after the first infusion at each dose level. For other infusions of CA4P limited sampling was performed. Samples were immediately placed on ice and protected from light. After centrifugation, plasma was removed with a pipet and transferred to a -20℃ freezer. Urine was also collected over 0 to 4 hours and 4 to 24 hours, the total volume was measured, and a 20-mL sample was stored at -20℃. Concentrations of CA4P, CA4, and the principal metabolite, CA4 glucuronide (CA4G), were measured using high-performance liquid chromatography.18 In brief, after addition of internal standard, the plasma was extracted with methanol, centrifuged, and the supernatant dried in a centrifugal evaporator. The sample was reconstituted and the CA4P, CA4, and CA4G were separated by reversed-phase ion-pair chromatography with fluorescence detection. Area under the curve (AUC) and the area under the first moment curve were calculated using the linear trapezoidal rule, and half-lives were calculated by weighted nonlinear least squares regression. For CA4P, clearance was calculated as dose/AUC, and the volume of distribution at steady-state (Vdss) was calculated as the product of the clearance and the ratio of the area under the first moment curve to the AUC.
RESULTS
Thirty-four patients were recruited onto the trial: 13 patients at the Hammersmith Hospital and 21 patients at Mount Vernon Hospital. Patient characteristics are listed in Table 1. All patients had measurable disease at baseline with masses at least 2 cm in diameter in all but one patient; however, eight patients discon- tinued study treatment before they could be evaluated for response. Five of the six patients with sarcomas had leiomyo- sarcomas. Doses were doubled for the first four dose levels (Table 2). At 40 mg/m2 one patient had grade 2 drug-related tumor pain, and dose escalations then proceeded by increments of 1.3 up to 114 mg/m2. Three patients were treated at two dose levels and three patients were treated at three dose levels. Intrapatient dose escalations were stopped at 88 mg/m2 when one patient had a DLT (vasovagal episode); the dose level was then expanded to six patients. At 114 mg/m2 two of six patients had DLTs (ataxia and motor neuropathy). Because one of these patients also had a DLT when re-treated at 88 mg/m2 and another patient had a DLT at 88 mg/m2, the lower dose of 68 mg/m2 was expanded to establish the MTD. An additional five patients were then treated at 52 mg/m2 to investigate further tumor blood flow effects at doses below the MTD.
Toxicity
All patients were assessable for toxicity; Table 3 lists the toxicities that were possibly, probably, or almost certainly drug related and only these are discussed below. The most common toxicities were cardiovascular, with tachycardia in 18 patients (53%), bradycardia in eight patients (24%), and hypertension in 12 patients (35%). However, only one patient with tachycardia
| Characteristic | No. of Patients |
|---|---|
| Assessable for response | 26 |
| Sex | |
| Male | 12 |
| Female | 22 |
| Age, years | |
| Mean | 52 |
| Range | 22-69 |
| WHO performance status | |
| 0 | 11 |
| 1 | 18 |
| 2 | 5 |
| Prior therapy | |
| Chemotherapy | 25 |
| Radiotherapy | 9 |
| Hormonal or biological therapy | 7 |
| Tumor type | |
| Soft tissue sarcoma | 6 |
| Colon | 5 |
| Kidney | 5 |
| Ovary | 5 |
| Liver | 2 |
| Rectum | 2 |
| Unknown primary | 2 |
| Adrenal | 1 |
| Breast | 1 |
| Lung | 1 |
| Melanoma | 1 |
| Esophagus | 1 |
| Osteosarcoma | 1 |
| Small intestine | 1 |
Abbreviation: WHO, World Health Organization.
and three patients with hypertension had toxicity higher than grade 1. During the study, additional cardiac monitoring was introduced because QTc interval prolongation had been reported in another phase I trial of CA4P. However, no ECG changes were seen up to 4 hours after infusion or in the 24-hour Holter monitoring performed in five patients.
Figure 1 illustrates the mean change in vital signs seen in eight patients treated at 5 to 40 mg/m2 and in 22 patients treated at 52 to 114 mg/m2. There were no significant changes in the group treated at 5 to 40 mg/m2. At the higher dose levels, blood pressure was significantly increased by a mean of 11 mmHg
| Dose Level (mg/m2) | No. of Patients (n = 34) | Number Escalated From Previous Dose | Total Number of Infusions | Number With DLTs |
|---|---|---|---|---|
| 5 | 3 | 0 | 8 | |
| 10 | 3 | 1 | 12 | |
| 20 | 3 | 2 | 9 | |
| 40 | 4 | 2 | 11 | |
| 52 | 8 | 1 | 27 | 1 |
| 68 | 10 | 2 | 52 | |
| 88 | 7 | 1 | 28 | 2* |
| 114 | 6 | 0 | 20 | 2* |
Abbreviations: CA4P, combretastatin A4 phosphate; DLTs, dose-limiting toxicities. *Only one patient had a dose reduction and had a DLT at both doses.
| Dose Level | 5-40 mg/m2 | 52 mg/m2 | 68 mg/m2 | 88-114 mg/m2 | ||||
|---|---|---|---|---|---|---|---|---|
| 1-2 | 3-4 | 1-2 | 3-4 | 1-2 | 3-4 | 1-2 | 3-4 | |
| Hemoglobin | 1 | 2 | 1 | 3 | ||||
| Lymphocytes | 2 | 1 | 6 | 2 | ||||
| Sinus bradycardia | 1 | 2 | 5 | |||||
| Sinus tachycardia | 5 | 4 | 10 | |||||
| Vasovagal episode | 1 | |||||||
| Hypertension | 1 | 2 | 3 | 6 | ||||
| Hypotension | 1 | 1 | 2 | 3 | ||||
| Bowel ischemia | 1 | |||||||
| Fatigue | 1 | 1 | 4 | 2 | ||||
| Sweating | 2 | 3 | 3 | |||||
| Diarrhea | 1 | 1 | 2 | |||||
| Nausea | 3 | 4 | ||||||
| Vomiting | 1 | 2 | 4 | |||||
| Ataxia | 1 | 1 | ||||||
| Dizziness or light headedness | 3 | 1 | 2 | |||||
| Neuropathy, motor | 1 | 1 | ||||||
| Neuropathy, sensory | 1 | 1 | 2 | |||||
| Vision | 1 | 2 | 1 | |||||
| Abdominal pain or cramping | 2 | 1 | 2 | 3 | ||||
| Headache | 1 | 1 | 4 | 2 | ||||
| Tumor pain | 1 | 4 | 2 | 4 | 1 | |||
| Dyspnea | 1 | 1 | ||||||
NOTE. The highest toxicity grade is counted for each patient at a particular dose. Abbreviation: CA4P, combretastatin A4 phosphate.
systolic (8%) and 8 mmHg diastolic (10%) 30 minutes to 1 hour after treatment (P < . 001, paired t test), and was associated with a 15% decrease in heart rate (P < . 001). Four hours after treatment systolic and diastolic blood pressure were significantly decreased by a mean of 8 and 6 mmHg (6% and 7%, respec- tively), respectively (P = . 07 and P = . 02, respectively), and heart rate was increased by 15% (P < . 001) to 98 beats/min. After 24 hours there was no significant difference in heart rate or blood pressure compared with pretreatment levels.
One 47-year-old male patient with a large left supraclavicular fossa nodal mass had a vasovagal episode 4.5 hours after his second infusion of 88 mg/m2 CA4P. He required morphine for abdominal pain 30 minutes after the infusion and at 80 minutes started vomiting, sweating, and became hypertensive for 30 minutes. His syncopal episode lasted 3 minutes and was accom- panied by incontinence and shaking. The episode was classified as a DLT. One patient had transient hypotension 4 to 7 hours after treatment at 68 mg/m2 CA4P and required brief intravenous fluid replacement, but this was not classified as a DLT.
One 63-year-old female patient with recurrent leiomyosar- coma in the pelvis died from small bowel ischemia after receiving her second dose of 52 mg/m2 CA4P as part of the final expansion of that dose level. She developed abdominal pain and vomiting 4 days after receiving CA4P and died 2 days later. Seventeen years earlier, she had received 50 Gy in 25 fractions of pelvic radiotherapy and had radiation proctitis proven by sigmoidoscopy in 1994. At postmortem there appeared to be small bowel ischemia confined to areas that were within the previous radiation field and close to the tumor that involved the mesentry of the affected bowel. This was classified as a DLT.
The next most common toxicity was pain in the site of a tumor mass, which was reported in 12 patients (35%) at doses from 40 to 114 mg/m2. The median time to onset was 40 minutes from start of the CA4P infusion (range, 2 minutes to 10 hours; duration, 1 to 180 minutes). However one patient who received 24 infusions of CA4P consistently experienced pain that recurred intermittently 24 to 36 hours after treatment and lasted longer than in other patients. Eight patients (24%) experienced abdom- inal or cramping pain that could not be localized to a known tumor mass. One of these patients died from small bowel ischemia, as discussed above.
Hematologic toxicity was mild and there was no drug-related neutropenia or thrombocytopenia. Nine patients (27%) experi- enced lymphopenia that was possibly or probably drug related, but only at dose levels from 52 to 88 mg/m2, which possibly indicates a relationship to the study drug. However, the level of lymphopenia was already low in five of these patients before treatment. Grade 1 or 2 anemia was observed in seven patients (20%) for most doses. Fatigue was seen in eight patients (23%) mostly at higher doses and after repeated infusions. Sweating was also reported in eight patients, of whom five also experi- enced pain. Mild headache was also recorded in eight patients.
Nausea and vomiting was each seen in seven patients (21%) at doses from 52 to 114 mg/m2, was usually mild, and sometimes started within minutes of beginning treatment with CA4P. It was easily controlled by standard antiemetics, which were not rou- tinely administered. Grade 1 and 2 diarrhea was reported in four patients but medication was only required for one patient.
Mild sensory neuropathy described as a tingling sensation in a variety of sites occurred in four patients (12%) over the whole
Dose levels 5-40 mg/m
2
Systolic BP
A
Diastolic BP
160
Pulse
Blood Pressure (mmHg)
Pulse rate (min -1)
140
120
100
80
60
infusion
40
0
20
40
60
80
100
120
140
160
180
200
220
240
260
Time (mins)
2
B
Dose Levels 52-114 mg/m
Systolic BP
Diastolic BP
160
Pulse
Blood Pressure (mmHg)
140
Pulse rate (min
120
100
80
60
infusion
40
0
20
40
60
80
100
120
140
160
180
200 22
240
20 24
0 26
Time (mins)
dose range. The neuropathy generally began within the first hour of starting CA4P treatment and resolved within 30 minutes. Three patients (9%) experienced visual toxicity. One patient had grade 2 blurred vision for 2 minutes that began 4 hours after her first infusion of 114 mg/m2; the visual toxicity did not recur after two additional infusions. Another patient who also received 114 mg/m2 had grade 2 flashing for 15 minutes that began 100 minutes after her first infusion of CA4P but did not occur after the next four infusions. The third patient experienced 30 minutes of grade 3 double vision that began 90 minutes after 88 mg/m2 CA4P; this patient previously suffered ataxia at 114 mg/m2.
Dose escalation stopped at 114 mg/m2 after two of six patients developed ataxia. One patient experienced unsteady gait (grade 2 ataxia) that lasted for 1 day and started approximately 15 hours after her only infusion of CA4P. The other patient experienced ataxia lasting for 8 hours, which started 22 hours after his first infusion of CA4P. One week later he was given 88 mg/m2 CA4P and developed grade 4 ataxia, which lasted 40 minutes and started 2 hours after administration of the CA4P. The second episode of ataxia was also associated with severe leg weakness and double vision mentioned above.
Two patients experienced dyspnea. The first patient had pre-existing dyspnea caused by metastatic leiomyosarcoma in-
volving more than half the right chest cavity. She experienced severe tumor pain 10 minutes after the infusion of 88 mg/m2 CA4P. Severe breathlessness developed 40 minutes after CA4P and only slightly improved over the following few days. ECGs performed from 1 to 24 hours after CA4P showed a positional shift of the heart but no ischemia. An MRI scan performed 4.5 hours after CA4P showed that there had been a dramatic enlargement of the tumor with mediastinal shift and move- ment of the heart, which the next day was shown to be normal by echocardiogram. The dyspnea was considered to be due to sudden edema or hemorrhage into the tumor. The change from pre-existing grade 1 dyspnea to grade 3 dyspnea after CA4P was not considered to be a DLT. The other patient developed dyspnea lasting up to 25 minutes starting 1 hour after the first two of six infusions of 114 mg/m2 CA4P. This patient had liver metastases recorded before starting treatment with CA4P, but had developed a pleural effusion and ascites after six infusions.
Response
There were no complete or partial responses. One patient with liver metastases from adrenocortical carcinoma had progression of one marker lesion after one cycle (three infusions) but wished to continue treatment. The sum of the product of the four marker lesions had decreased by 51% from baseline after cycle 3, by 54% after cycle 4, by 48% after cycle 5, and by 20% after cycle 7, which was 38 weeks after starting treatment with CA4P. Her pretreatment and after fourth cycle T2-weighted MRI images are shown in Figure 2. The reduction in tumor size was associated with a reduction in the levels of adrenal androgens and cortico- steroids measured on 24-hour urine collection. Grossly elevated levels of 5ß-etiocholanolone, dehydroepiandrosterone, 16a-hy- droxy-DHA, 11ß-hydroxy androsterone, 11ß-hydroxy etiochol- anolone, and 5ß-tetrahydrocortisol decreased to less than half their pretreatment levels after four cycles of CA4P but were elevated again after seven cycles. She had documented progres- sion of disease 9 months after beginning treatment, when a new lesion was noted, at which time her adrenal hormone levels had increased. Two patients had stable disease after two cycles of three infusions, but all of these had progressed after three or four cycles. Nine patients had evidence of disease progression within the first cycle of treatment.
Pharmacokinetics
As the samples were analyzed during the trial, it became apparent that in seven patients, a limited number of samples immediately after the infusion had been taken without EDTA, resulting in variable ex vivo conversion of CA4P to CA4. These patients were excluded from the data analysis, and a total of 33 profiles from 27 patients were suitable for evaluation.
CA4P was rapidly dephosphorylated to the active CA4 with a half-life of a few minutes. At the lower doses, because of the rapid clearance and the sensitivity limits of the assay, only a distribution phase could be seen, but above 20 mg/m2, a beta elimination phase could also be determined. CA4 was further
A
A
A
10-FEB-2000
10-FLE-2000
[TTTTTTTTTI
R
-160.7
150.7
E
1032 357
A
A
10-FEB-2000
10-FEB-2000
R
140.7
SP
130.7
B
A
A
11-JUL-2000
11-JUL-2009
111.4
101.4
E
恐
. :
A
A
11-JUL-2000
11-JUL-2000
1.4
.4
metabolized at a slower rate to the glucuronide (CA4G). A typical plasma profile seen after 68 mg/m2 is shown in Figure 3.
In all but one patient, the peak CA4 concentration was seen at the end of the infusion, whereas CA4G concentrations were at their maximum 15 minutes after the end of the infusion. CA4 AUC and peak concentration (Cmax) increased in a linear fashion with doses from 0.169 pmol · h/L and 0.407 µmol/L at 5 mg/m2 to 3.29 pmol · h/L and 4.46 µmol/L at 114 mg/m2 (Fig 4).
Clearance of CA4P for all patients was 30.4 + 14.8 L . h-1 and no dependence on either dose or body-surface area was observed. There was a trend for Vass (mean for all patients 4.64 ± 1.72 L) to increase with dose, but this failed to reach significance (r = . 328; P = . 063). Urinary excretion of CA4 was exclusively as the glucuronide. The mean recovery of adminis- tered drug in all patients was 60% ± 14% (standard deviation). Pharmacokinetic parameters are summarized in Table 4 for eight of 10 patients who received 68 mg/m2 and for six of eight patients who received 52 mg/m2; these doses are being consid- ered for the phase II dose.
^ CA4G
CA4
o CA4P
Concentration (umol.[‘1)
10
1
0.1
0.01
0
5
10
15
Time
(Hours)
20
25
DISCUSSION
There is considerable current interest in agents that can interfere with tumor blood supply. The highest profile has been achieved by drugs that oppose the angiogenic substances re- leased by tumor cells, the so-called antiangiogenic agents. These
A
7
6
AUC (umol.h.l”1)
5
4
3
2
1
0
0
20
40
60
80
100
120
Dose (mg/m2)
B
6
5
peak (umol.l’1)
4
3
2
1
0
0
20
40
60
80
100
120
Dose (mg/m2)
| Parameter | CA4P | CA4 | CA4G | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 52 mg/m2 | 68 mg/m2 | 52 mg/m2 | 68 mg/m2 | 52 mg/m2 | 68 mg/m2 | |||||||
| Error | SD | Error | SD | Error | SD | Error | SD | Error | SD | Error | SD | |
| +1/2a, hours | 0.103 | 0.040 | 0.078 | 0.027 | 0.221 | 0.127 | 0.260 | 0.120 | 0.440 | 0.163 | 0.354 | 0.095 |
| +1/2B, hours | 0.489 | 0.420 | 0.301 | 0.159 | 2.23 | 1.07 | 1.93 | 0.58 | 4.49 | 1.68 | 2.92 | 0.71 |
| AUC, pmol . h . L-1 | 9.07 | 3.76 | 12.0 | 3.7 | 2.19 | 0.58 | 2.33 | 0.55 | 33.9 | 14.8 | 37.7 | 18.2 |
| Peak, umol · L-1 | 30.3 | 9.95 | 46.3 | 12.5 | 1.89 | 0.87 | 2.26 | 0.62 | 8.94 | 2.51 | 13.4 | 1.45 |
| Clearance, L . h-1 | 29.5 | 13.7 | 25.3 | 9.9 | - | - | - | - | ||||
| Vdss/ L | 4.77 | 1.60 | 3.95 | 1.39 | - | - | - | - | ||||
Abbreviations: CA4P, combretastatin A4 phosphate; CA4, combretastatin A4; CA4G, combretastatin glucuronide; SD, standard deviation; t1/2a, half life alpha; t1/2B, half life beta; AUC, area under the curve; Vass, volume of distribution at steady-state.
include drugs that block matrix breakdown by metalloprotein- ases, drugs that inhibit endothelial cells directly (such as thalid- omide, squalamine, and endostatin), and drugs that block acti- vators of angiogenesis such as vascular endothelial growth factor.19,20 Vascular targeting is a different approach, because the aim is not prevention of the growth of a tumor’s vessels, but rather the rapid and selective destruction of the vasculature already established by the tumor. Tumor vasculature differs from normal vasculature in several ways. It has on average a prolif- eration rate that is 50-fold higher than normal vasculature and is therefore less mature.21 It is structurally different, with vessels that are thin-walled, tortuous, and fragile, and that lack smooth muscle coats and innervation.22 Tumor vasculature also may overexpress certain surface antigens such as vascular endothelial growth factor or integrins.23
The proposal that differences in tumor vasculature might lend themselves to therapeutic attack was first suggested by Denekamp in 1982.24 Each vessel supplies many thousands of tumor cells, so damage to a small section of vessel that causes occlusion will cause large-scale tumor cell death downstream. There should not be delivery problems and because the endo- thelial cells are not malignant, they should not have the capacity to mutate and develop drug resistance. Although vascular tar- geting agents might be capable of inducing ischemic necrosis to the center of tumors supplied by the tumor vasculature, there is likely to be a viable peripheral rim of tumor cells capable of rapid repopulation. This makes it likely that tumor response may not be seen after these agents are given alone. It was therefore considered essential to measure tumor blood flow after CA4P. The reduction that was demonstrated in the accompanying articles supports combining CA4P with cytotoxic agents, radio- therapy, or radiolabeled antibodies in future studies because the combinations have activity in animal models.7,8,25,26
CA4P showed a different spectrum of toxicities from conven- tional chemotherapy, with no neutropenia or thrombocytopenia, but a mild lymphocytopenia at higher doses. It was generally well tolerated, with only mild nausea or vomiting that was easily controlled by antiemetics and could probably be prevented by premedication. DLT was reversible ataxia at 114 mg/m2. Al- though tubulin-binding agents are known to cause neurologic toxicity and this was predicted in the dog toxicology studies, the etiology is unclear. The only fatal DLT was small bowel ischemia at 52 mg/m2. This was only seen in bowel that was
compromised by tumor and previous radical radiotherapy. The fact that the bowel had been previously shown to be affected by radiotherapy indicates that if normal tissue blood vessels are already damaged, they are at increased risk of causing irrevers- ible ischemia after CA4P. Prior radical radiotherapy that could be associated with late normal tissue damage should be consid- ered a relative contraindication to the administration of CA4P.
Adverse events were often of an acute and transient nature, which created difficulties in grading because the toxicity criteria used were largely designed to detect chronic changes.
There was a typical course of events for each patient, reliably repeated after each course. There were no cumulative effects seen after repeated infusions. Pain at the sites of tumor either increased or developed in more than one third of patients at dose levels between 40 and 114 mg/m2. It is not clear whether this pain is related to tumor ischemia or to acute swelling of the tumor caused by alteration in endothelial cell permeability after CA4P. Enlargement of a large intrathoracic sarcoma over 4 hours was clearly documented in one patient, and enlargement of tumor masses was noted in other patients within hours of receiving CA4P. Swelling of neck nodal masses could be the explanation of the vasovagal episode that was classified as a DLT. Tumor pain was not considered a DLT because it could be controlled by analgesics and was tolerated by patients, who saw it as a possible manifestation of efficacy.
Cardiovascular adverse events were confined to changes in pulse and blood pressure in this study, with no cardiac events and no evidence of changes in ECG parameters. At doses above 40 mg/m2 blood pressure was significantly increased within the first hour after infusion and was associated with a decrease in heart rate. This coincides with the significant decrease in cardiac output demonstrated by PET at this time.16 No significant difference was noted in stroke volume, indicating that the increase in blood pressure and decrease in heart rate was a compensatory response to an increase in peripheral resistance caused by CA4P, which was clearly documented in animal studies.9 Four hours after treatment there was a significant decrease in blood pressure back to pretreatment levels and an increase in heart rate. These changes coincide with the timing of the vasovagal episode and indicate that patients need monitoring for several hours after higher doses of CA4P. Normal tissue DCE-MRI scans at 4 hours showed no signif- icant mean change in parameters related to blood flow,
although significant decreases in kidney, liver, or spleen parameters were seen in four patients.15
This trial used an accelerated titration design and intrapatient dose escalation. Eight patients were initially treated at doses below 52 mg/m2, which is the lowest dose at which changes in parameters associated with blood flow reduction were seen15; one of these patients was escalated to 52 mg/m2. If one accepts that 52 mg/m2 was the dose that needed to be reached quickly, this design fulfilled that requirement, with only seven patients receiving doses considered to be inactive. Doses were safely doubled from three dose levels and the only increase in toxicity seen in more than one of the six patients after their dose was escalated was tumor pain.
The rapid activation of the phosphate prodrug to the active CA4 was expected from published work with other phosphate ester prodrugs such as etoposide phosphate and our preclinical studies with CA4P. The additional extensive metabolism to the glucuronide results in mean CA4 AUCO-24 hours and Cmax in eight patients at 68 mg/m2 of 2.33 umol . h/L and 2.26 µmol/L, respectively. Comparable figures for 52 mg/m2 are 2.19 pmol · h/L and 1.89 µmol/L, respectively. This is somewhat lower than the figures in the mouse bearing the carcinoma CaNT (AUC and
Cmax 5.8 pmol · h/L and 9.8 mmol/L, respectively) at an active dose of 25 mg/kg. However, at this dose more than 40% vascular volume reduction is seen in this responsive tumor model (S. Hill, personal communication, July 2002), and in this clinical trial these concentrations resulted in a significant decrease in mean kinetic parameters in tumor by DCE-MRI scans. The accom- panying paper also shows that there was a significant corre- lation between the AUC of CA4 and relative change in DCE-MRI kinetic parameters.15
On the basis of just toxicity criteria, the dose recommended for phase II trials would be 68 mg/m2 CA4P because this dose was well tolerated in nine of 10 patients. However, for a vascular targeting agent, a lower dose could be considered for phase II trials if significant reductions in parameters associ- ated with tumor blood flow were seen at that dose. The accompanying articles show significant reductions at 52 and 68 mg/m2 by both PET and DCE-MRI techniques.15,16 We therefore recommend that the phase II dose should be in the dose range of 52 to 68 mg/m2.
ACKNOWLEDGMENT
We are grateful for the assistance of the research nurses Jane Boxall and Cathy Purnell, and the team at the Drug Development Office at CRUK.
REFERENCES
1. Pettit GR, Cragg GM, Herald DL, et al: Isolation and structure of combretastatin. Can J Chem 60:1374-1376, 1982
2. Pettit GR, Singh SB, Hamel E, et al: Isolation and structure of the strong growth and tubulin inhibitor combretastatin A-4. Experentia 45:209-211, 1989
3. Pettit GR, Temple C Jr, Narayanan VL, et al: Antineoplastic agents 322. synthesis of combretastatin A-4 prodrugs. Anticancer Drug Des 10:299-309, 1995
4. Hamel E, Lin CM: Interactions of combretastatin, a new plant-derived anti-mitotic agent, with tubulin. Biochem Pharmacol 32:3864-3867, 1983
5. Chaplin DJ, Pettit GR, Parkins CS, et al: Antivascular approaches to solid tumour therapy: Evaluation of tubulin binding agents. Br J Cancer 74:S86-S88, 1996
6. Dark GG, Hill SA, Prise VE, et al: Combretastatin A-4, an agent that displays potent and selective toxicity toward tumour vasculature. Cancer Res 57:1829-1834, 1997
7. Chaplin DJ, Pettit GR, Hill SA: Anti-vascular approaches to solid tumour therapy: Evaluation of combretastatin A4 phosphate. Anticancer Res 19:189-195, 1999
8. Pedley R, Hill S, Boxer G, et al: Eradication of colorectal xenografts by combined radioimmunotherapy and combretastatin A-4 3-O-phosphate. Can- cer Res 61:4716-4722, 2001
9. Tozer GM, Prise VE, Wilson J, et al: Combretastatin A-4 phosphate as a tumor vascular-targeting agent: Early effects in tumors and normal tissues. Cancer Res 59:1626-1634, 1999
10. Grosios K, Holwell SE, McGown AT, et al: In vivo and in vitro evaluation of combretastatin A-4 and its sodium phosphate prodrug. Br J Cancer 81:1318-1327, 1999
11. Watts ME, Woodcock M, Arnold S, et al: Effects of novel and conventional anti-cancer agents on human endothelial permeability: Influ- ence of tumour secreted factors. Anticancer Res 17:71-75, 1997
12. Galbraith SM, Chaplin DJ, Lee F, et al: Effects of combretastatin A4 phosphate on endothelial cell morphology in vitro and relationship to tumour vascular targeting activity in vivo. Anticancer Res 21:93-102, 2001
13. Dowlati A, Robertson K, Cooney M, et al: A phase I pharmacokinetic and translational study of the novel vascular targeting agent combretastatin A-4 phosphate on a single-dose intravenous schedule in patients with advanced cancer. Cancer Res 62:3408-3416, 2002
14. Stevenson JP, Gallagher M, Sun W, et al: Phase I/pharmacokinetic trial of the endothelial toxin combretastatin A4P (CA4P) administered as an IV bolus on a daily x 5 schedule every 21 days. Proceedings of the 91st annual meeting of the Am Assoc Cancer Res 2000:3469
15. Galbraith SM, Maxwell RJ, Lodge MA: Combretastatin A4 phosphate has tumor antivascular activity in rat and man as demonstrated by dynamic magnetic resonance imaging. J Clin Oncol 21:2831-2842, 2003
16. Anderson HL, Yap JT, Miller MP: Assessment of pharmacodynamic vascular response in a phase I trial of combretastatin A4 phosphate. J Clin Oncol 21:2823-2830, 2003
17. Simon R, Freidlin B, Rubinstein L, et al: Accelerated titration designs for phase I clinical trials in oncology. J Natl Cancer Inst 89:1138-1147, 1997
18. Stratford MR, Dennis MF: Determination of combretastatin A-4 and its phosphate ester pro-drug in plasma by high-performance liquid chroma- tography. J Chromatogr B Biomed Sci Appl 721:77-85, 1999
19. O’Reilly MS, Boehm T, Shing Y, et al: Endostatin: An endogenous inhibitor of angiogenesis and tumor growth. Cell 88:277-285, 1997
20. Zhu Z, Witte L: Inhibition of tumor growth and metastasis by targeting tumor-associated angiogenesis with antagonists to the receptors of vascular endothelial growth factor. Invest New Drugs 17:195-212, 1999
21. Kakolyris S, Fox S, Koukourakis M, et al: Relationship of vascular maturation in breast cancer blood vessels to vascular density and metastasis, assessed by expression of a novel basement membrane component, LH39. Br J Cancer 82:844-851, 2000
22. Denekamp J: The current status of targeting tumour vasculature as a means of cancer therapy: An overview. Int J Radiat Biol 60:401-408, 1991
23. Bloemendal H, Logtenberg T, Voest E: New strategies in anti- vascular cancer therapy. Eur J Clin Invest 29:802-809, 1999
24. Denekamp J: Endothelial cell proliferation as a novel approach to targeting tumour therapy. Br J Cancer 45:136-139, 1982
25. Grosios K, Loadman PM, Swaine DJ, et al: Combination chemother- apy with combretastatin A-4 phosphate and 5-fluorouracil in an experimental murine colon adenocarcinoma. Anticancer Res 20:229-233, 2000
26. Li L, Rojiani A, Siemann D: Targeting the tumor vasculature with combretastatin A-4 disodium phosphate: Effects on radiation therapy. Int J Radiat Oncol Biol Phys 42:899-903, 1998