Phase I and Pharmacokinetic Study of E7070, a Chloroindolyl-Sulfonamide Anticancer Agent, Administered on a Weekly Schedule to Patients with Solid Tumors

E7070 is a novel chloroindolyl-sulfonamide anticancer agent with a wide range of activity in human tumor cell lines (1). In the A549 human NSCLC² cell line treatment with E7070 produced a profound arrest at the G₁-S boundary accompanied by hypophosphorylation of the retinoblastoma protein and a reduction in the expression of cyclins A, B1, and cyclin-dependent kinase 2. At higher concentrations, E7070 exposure was associated with up-regulation of p53 and p21 and subsequent apoptosis (2). E7070 treatment produced cures in up to 80% of human colorectal cancer cell line HCT116 and lung cancer LX-1 xenografts. Strong evidence of schedule dependency was observed in all preclinical models with repeated daily dosing showing greatest efficacy (1). In the NCI COMPARE screening program E7070 displayed low correlation coefficients when compared with other anticancer drugs, suggesting a unique mechanism of action (1). E7070 shows similarities with chloroquinoxaline-sulfonamide, a compound developed by the NCI that was shown to exhibit a hazardous toxicity profile characterized by dose-limiting hypoglycemia and cardiac tachyarrhythmias that prevented additional clinical development. Preclinical safety studies reported both a transient reduction in cardiac contractility and a minor prolongation of the corrected QT interval occurring 30–60 min after a 5-min infusion of the MTD of E7070. Similar changes were not seen after administration of the drug over 1 h (E7070 Investigator’s Brochure, version 6). Treatment with E7070 was also associated with a diuretic effect which was thought to be related to inhibition of carbonic anhydrase. This enzyme is intimately involved in the production of aqueous humor and is thus involved in maintaining intraocular pressure (E7070 Investigator’s Brochure, version 6). Although the mechanism of action of chloroquinoxaline-sulfonamide and E7070 are thought to be similar, E7070 displayed antiproliferative effects that were ∼10 times more potent in human colorectal and NSCLC models. Hypophosphorylation of the retinoblastoma protein corresponding to induction of G₁-S arrest and apoptosis have been observed in samples of head and neck tumors treated with the agent. Apoptosis and cell cycle disruption have also been documented from bronchial brushings taken from patients with endobronchial NSCLC receiving E7070 (3).

Four separate Phase I studies of E7070 have been performed by the European Organization for Research and Treatment of Cancer-Early Clinical Studies Group/New Drug Development Program using the following schedules: a single i.v. infusion every 3 weeks; a daily i.v. infusion on 5 consecutive days every 3 weeks; a weekly i.v. infusion given on 4 consecutive weeks repeated six weekly; and a continuous i.v. infusion for 5 days every 3 weeks. Final data on the once every 3 week and daily times five schedule (4, 5) and preliminary results on the other studies have been reported (6, 7, 8).

The objectives of this study were (a) to determine the MTD and RD for additional evaluation (b), to define the DLT (c), to study the pharmacokinetics of E7070, and (d) to seek preliminary evidence of antitumor activity using a weekly schedule of administration.

Eligibility criteria included patients with a histologically confirmed diagnosis of a solid tumor who had failed previous therapy or were considered not amenable to established forms of treatment, age ≥ 18 years, WHO performance score ≤ 2, no prior systemic treatment or radiotherapy within 4 weeks (6 weeks for nitrosoureas, mitomycin C, and extensive radiotherapy), adequate bone marrow function as defined by neutrophils ≥ 1.5 × 10⁹/liter, platelets ≥ 100 × 10⁹/liter and hemoglobin ≥ 9 g/dl, adequate hepatic function as defined by bilirubin < 25 mmol/liter and aspartate aminotransferase/alanine aminotransferase ≤ 2.5 times the ULN (unless in the presence of known liver metastases in which case aspartate aminotransferase/alanine aminotransferase have to be less than or equal to five times the ULN) and adequate renal function as defined by serum creatinine < 120 mmol/liter (1.4 mg/dl), no concurrent treatment with sulfonylurea hypoglycemics or antiarrhythmic agents, absence of glaucoma, and written informed consent. Patients must have had a psychosocial state compatible with participation, including no history of a previous malignancy, alcoholism, drug addiction, or psychotic disorders. Patients were excluded if they were pregnant or breast feeding (men and women of childbearing potential were required to use a reliable method of contraception) or had symptomatic brain metastases/leptomeningeal tumor involvement. The study was performed after approval by the local ethics committees of the three participating institutions.

E7070 was manufactured by EISAI (Tuskuba Laboratories, Tuskuba, Japan) and formulated for i.v. injection at Ben Venue Laboratories, Inc. (Bedford). Drug was supplied in ready-to-use 15-ml glass vials containing 100 mg of E7070 as the anhydrous-free base equivalent, which was diluted with 5 ml of water for injection to obtain a final concentration of 18.35 mg/ml. The appropriate volume of this stock solution was added to isotonic saline to yield the required dose. E7070 was administered in 500 ml of normal saline as a 1 h i.v. infusion on 4 consecutive weeks. Treatment was repeated six weekly. Prophylactic antiemetics were not administered routinely.

On the basis of 1 of 10 of the rat MTD, a starting dose of 40 mg/m²/week of E7070 was selected. Dose escalation was performed in cohorts of 3–6 patients according to a Fibonacci-like schedule. There was no intrasubject dose escalation foreseen. Doses were escalated in 100–50% increments in the presence of no or grade 1 toxicity and in 33–20% increments in the presence of grade 2 or more toxicity (except for alopecia, anemia, or untreated nausea/vomiting). The results of ongoing pharmacokinetic analyses were also considered in identifying subsequent dose levels. The absolute doses investigated were 40, 80, 160, 320, 400, and 500 mg/m²/week. The MTD was defined as the dose at which the DLT defined as grade 4 hematological toxicity or grade 3 or more nonhematological toxicity directly related to E7070 occurred during the first 6-week cycle of E7070 in ≥33% of patients. To establish the MTD, patients leaving the study before completing the first 6-week cycle because of early progressive disease were replaced. The RD was defined as one dose level below the MTD.

Dose escalation occurred in two overlapping patient cohorts—firstly, in an unselected patient population that met the eligibility criteria described above (group A), and secondly, in an additional group of less heavily pretreated patients who were expected to tolerate a higher dose of E7070 (group B). To be included in group B, patients should not have received more than three previous lines of chemotherapy, prior treatment with mitomycin C, nitrosoureas or high dose chemotherapy with or without stem cell support, and must have had adequate hepatic function (liver function tests within 2.5 times the ULN, regardless of liver metastases), <50% liver replacement by tumor on imaging, and no evidence of ascites/pleural effusion.

To proceed to a higher dose level, at least 2 patients must have completed their first cycle of E7070 at the previous dose level without evidence of DLT, and the third patient must also have received the first two doses of E7070 uneventfully. Although in the absence of progressive disease or significant toxicity, patients could continue E7070, existing eligibility criteria (performance status, hematological, hepatic and renal function) had to be met before each cycle.

Complete medical history, physical examination, WHO performance status, biochemical profile, full blood count, and urinalysis were performed at baseline and repeated weekly. An ECG and a chest X-ray were obtained within 14 days before receiving E7070. Toxicity was evaluated by clinical and biological examination on a weekly basis and graded using the NCI-Common Toxicity Criteria, version 2.0 (NCI Division of Cancer Treatment Guidelines for Reporting of Adverse Drug Reactions). Additional laboratory tests and/or an increase in the frequency of observations were permitted to document acute drug-related toxicity until recovery. Patients were evaluable for toxicity if they had received at least one dose of E7070. The evaluation period included observations before recovery from all acute toxicities associated with the first cycle of E7070.

Tumor assessments were performed at six weekly intervals. To be evaluable for efficacy, patients must have undergone at least two tumor assessments at least 6 weeks apart. Response evaluation was performed according to the standard WHO tumor response criteria.

Intraocular pressure was measured within 2 weeks before the first dose of E7070, and additional measurements were planned only if clinical signs of visual impairment developed, e.g., loss of visual acuity and ophthalmic pain. Twenty-four-h continuous cardiac monitoring was performed before and after the first dose of E7070 (week 1). Traces were reviewed centrally at Hertford Medical Limited (Hertford, Hertfordshire, United Kingdom). Cardiac monitoring was repeated around subsequent courses only if a significant arrhythmia was detected during the first infusion. A 12-lead ECG was performed before starting treatment and at the end of each dose of E7070 throughout the first cycle of treatment. If the QTc interval (according to Bazett’s correction factor) was prolonged, further 12-lead ECG assessments were performed 6 and 12 h after the infusion and then at 12-h intervals until the QTc interval normalized. In the case of a documented increase in the QTc interval ECGs were repeated in subsequent courses of E7070.

Blood glucose was monitored before the first dose of E7070 and every hour for 4 h after the start of the first infusion. Blood glucose was monitored during subsequent cycles only if hypoglycemia was detected after the first dose of E7070. Creatinine clearance was repeated if serum biochemistry revealed an increase in serum creatinine > grade 1 or >1.5 times the ULN.

Samples of blood and urine were obtained from all patients after the first and fourth dose of E7070 during the first cycle of treatment. Blood samples were collected immediately before each infusion, 30 min after the start of the infusion, at the end of the infusion, as well as at 10 min, 30 min, 1, 2, 4, 6, 8, 12, 24, 36, 48, 72, and 96 h after the end of the infusion. In addition trough blood samples were collected before the second and third infusion. For the fourth infusion, sampling was identical to that performed around the first infusion. A pretreatment urine sample and 24-h collection in three 8-h aliquots were obtained after the first dose of E7070. Blood and urine samples were processed and analyzed as described previously (4).

Pharmacokinetic parameters were determined using standard noncompartmental analysis. Population pharmacokinetics of this study were performed by NOTOX (Pharmacokinetic Report, NOTOX Protocol No. 220567) and within the frame of a population pharmacokinetic model designed and validated to describe the pharmacokinetics of E7070 at all four treatment schedules tested to identify the possible influence of patient characteristics on the pharmacokinetic parameters (9).

Thirty-seven patients (group A) were entered into the study from three different centers to determine the overall MTD. One patient died before receiving E7070. Patient characteristics are detailed in Table 1. One patient did not meet the eligibility criteria having received chemotherapy within 4 weeks of entering the study but was included in the analysis. All patients who received E7070 were evaluated for safety. The MTD for the initial patient cohort, i.e., group A was established as 500 mg/m²/week. Additional dose escalation was intended in group B, which comprised 9 prospectively recruited patients and an additional 9 patients from group A who also met the additional eligibility criteria defined above and were identified retrospectively (Table 1). There were two protocol violators. One patient was enrolled despite receiving tamoxifen within 4 weeks of entering the study and another patient had a grade 1 elevation of serum creatinine at screening. Therefore, in group B, 17 patients were evaluated for safety and 16 patients for efficacy. The MTD in group B was also identified as 500 mg/m²/week. The median number of cycles delivered to the groups A and B were 3 (range, 0–21) and 2 (range, 1–6), respectively. In both cases, the DLTs were reversible neutropenia and thrombocytopenia.

A total of 132 cycles of E7070 was administered. According to the protocol, only toxicities occurring during the first cycle, i.e., within the first 6 weeks from the start of therapy, were considered in establishing the MTD. The dose escalation schedule and treatment-related grade 3/4 toxicities in the two patient cohorts are summarized in Table 2A for group A and Table 2B for group B.

The first dose level of 40 mg/m²/week was expanded to 6 patients because the first patient with a transitional cell carcinoma of the bladder and a history of diabetes mellitus, but who was not taking any antidiabetic medication at study entry, experienced grade 3 hyperglycemia during the first cycle of treatment. This event was specified as possibly related to E7070. No additional DLTs were recorded among the 3 additional patients recruited at this dose level. Similarly, no DLTs were recorded among three patients treated at 80 mg/m²/week. One patient treated at a dose of 160 mg/m²/week experienced rigors after repeated doses of E7070, but these were not considered to be a DLT, and this cohort was not expanded. An additional 3 patients were treated uneventfully at a dose of 320 mg/m²/week. The dose was escalated to 500 mg/m²/week where 1 of 3 evaluable patients experienced a DLT with grade 4 neutropenia and thrombocytopenia. Three additional patients were entered at this dose level, 2 of whom also experienced DLTs, including grade 4 neutropenia and thrombocytopenia and grade 3 stomatitis in one patient and in another grade 3 neutropenia, which was classified as DLT as it necessitated a dose reduction after the second dose of the first cycle to continue therapy. The time to onset of grade 3/4 neutropenia and thrombocytopenia in these patients was between 21 and 28 and 15 and 21 days with a duration of 3–12 and 2 and 6 days, respectively. Consequently, the dose was de-escalated to 400 mg/m²/week. At this level, 1 patient developed grade 4 neutropenia. An additional patient suffered grade 4 diarrhea, which was initially considered as possibly related to E7070, triggering an additional dose reduction to 320 mg/m²/week. Once again, at this dose level, no DLTs were observed. Upon review of the clinical information whereby the diarrhea was found to be clearly a consequence of a preexisting postgastrectomy “dumping syndrome,” the relationship between this toxicity and E7070 treatment was changed to unrelated, and the dose was then again escalated to 400 mg/m²/week. A total of 7 additional patients were treated to provide an additional 4 evaluable patients, which when added to the information obtained from 2 previously treated patients, gave a cohort of 6 evaluable patients at this dose level. Of these 6, 1 patient experienced grade 4 neutropenia. Although an additional patient treated as part of this cohort experienced grade 3 dehydration, this was considered to be the result of underlying disease progression with increasing drowsiness because of opiate usage neglecting additional fluid intake rather than related to treatment with E7070. Thus, 500 mg/m²/week was considered to be the MTD and 400 mg/m²/week the RD for this schedule in an unselected cohort of patients (Table 2A).

To determine whether the RD of 400 mg/m²/week could be safely escalated in a less heavily pretreated group of patients with adequate hepatic function and restricted tumor bulk, an additional dose escalation was performed in selected patients (group B). Nine patients who had not previously received treatment with E7070 and met the more stringent eligibility criteria described above were recruited into group B. In addition, an additional 9 evaluable patients from group A who met these criteria for group B were identified retrospectively (Table 2B). Because 1 of 3 patients who were treated at the 400 mg/m²/week dose level experienced a DLT, an additional 4 previously untreated patients were entered at this level. None of these 3 additional patients who were evaluable for the assessment of MTD experienced a DLT. The second patient in group B treated at a dose of 500 mg/m²/week developed sepsis in the presence of a normal neutrophil count while severely thrombocytopenic and died on day 10 of the first treatment cycle. Immediately before the second dose of E7070 on day 8, this patient’s blood count was as follows: hemoglobin 9.6 g/dl; neutrophils 6.2 × 10⁹/liter; and platelets 104 × 10⁹/liter. Two days later at the time of death, his unsupported platelet count had dropped to 15 × 10⁹/liter, whereas the remainder of this blood count remained unchanged. Several additional side effects of grade 3 severity considered to be DLTs were also observed in this patient. Rather than add additional patients at this dose level, 3 patients who had been treated with 500 mg/m²/week as part of the group A dose escalation were included in the evaluation of the group B 500 mg/m²/week dose cohort. In 1 of these patients, neutropenia, although grade 3 only, was classified to be a DLT because it necessitated a dose reduction after the second dose of the first cycle to continue therapy with E7070. Thus, in total, 2 of 5 patients treated at a dose of 500 mg/m²/week were considered to have experienced DLTs, and no additional exploration of this dose level was performed. The time to onset of grade 3 and 4 neutropenia and thrombocytopenia was between 21 and 28 and 9 days with a duration of 3–8 and 2 days, respectively. Thus, as with group A, the MTD for group B was defined as 500 mg/m²/week and the RD as 400 mg/m²/week.

In general, all toxicities were reversible and noncumulative. Common treatment-related toxicities for both groups are detailed in Table 3. In this study, a total of 6 evaluable patients in group A was treated with the RD of 400 mg/m²/week, and only 1 of those 6 (17%) required a dose reduction for myelosuppression during the first cycle, resulting in a single dose reduction of 15 cycles (7%; Table 2A). In group B, both grade 4 neutropenia and thrombocytopenia were recorded in a single additional patient at cycle 1 at the RD of 400 mg/m²/week. Together with a patient of group A of the same dose level who was retrospectively considered also for group B, 2 of 12 evaluable patients (17%) with a DLT in form of myelosuppression in 3 of 24 cycles (13%) were documented (Table 2B). One patient of group B withdrew from the study because of myelosuppression after the second dose of E7070 in cycle 2. The single death attributable to sepsis was considered possibly related to treatment with E7070. This patient developed grade 3 epistaxis and melena in the presence of grade 4 thrombocytopenia 2 days after his second dose of E7070 (day 10). In addition, he experienced grade 3 nausea, diarrhea, fever, and fatigue, which were all considered possibly related to E7070. One patient treated at the starting dose level of 40 mg/m²/week experienced a minor response and underwent a gradual dose escalation to 80, 160, and 320 mg/m²/week at cycles 9, 10–13, and 14, respectively. Grade 4 neutropenia occurred after the second dose of E7070 in cycle 14 (19 months into treatment), and an additional 7 cycles of E7070 were administered at a reduced dose of 160 mg/m²/week.

There were no symptoms suggestive of changes in intraocular pressure. Therefore, repeat measurements were not performed. Holter monitor traces blood pressure and heart rate measurements were unremarkable.

The pharmacokinetic parameters of E7070 were determined using standard noncompartmental analysis (Table 4, Fig. 1). The intersubject variability in results at each dose level was low. The pharmacokinetics of E7070 were nonlinear at doses >80 mg/m² with a disproportional increase in exposure to E7070 assessed as the dose normalized AUC at higher dose levels. The effect of administered dose on dose-normalized AUC was highly significant (P < 0.001) for both AUC_{0–24 h} and AUC_0–∞. There was no significant difference between weeks 1 and 4 for any of the pharmacokinetic parameters, indicating that there was no evidence of drug accumulation. In contrast to the results for the AUC, the dose-normalized plasma concentration at the end of the infusion corrected for administered dose (C_end) did not vary significantly with administered dose (P = 0.30). The effect of administered dose on measures of elimination (t_1/2, MRT, Cl) was inconsistent. Although t_1/2 did not vary significantly between dose levels, there was a significant effect of administered dose on both MRT (P = 0.006) and Cl (P < 0.001). The t_1/2 of E7070 was ∼20 h, although there was considerable interpatient variation (range, 8–73 h). The drug was predominantly eliminated by metabolism with the renal Cl of unchanged E7070 representing <1% of administered dose.

The pharmacokinetics of E7070 were studied in 5 patients during two cycles of treatment. In all cases, the AUC values obtained from the second cycle of treatment were comparable with those seen in the first. The patient who underwent dose escalation to 80, 160, and 320 mg/m²/week and finally dose reduction to 160 mg/m²/week underwent pharmacokinetic analysis at the following dose levels: 40 mg/m²/week, cycle 1 (weeks 1–4); 320 mg/m²/week, cycle 14 (week 1); and 160 mg/m²/week, cycle 18 (weeks 1 and 4; Fig. 2, A and B).

A nonlinear pharmacokinetic model, including three dispositional compartments and a saturable elimination pathway describing the plasma concentration-time profiles of E7070 after single dose i.v. infusion, was fitted to the data using a population approach and revealed that it fitted individual profiles well. The pharmacokinetics of E7070 infused with a constant rate was nonlinear and the decline from linearity became evident at dosages exceeding 80 mg/m² (Fig. 3). Most of the parameters were highly variable across the studied patient population. The body surface area turned out to be the demographic covariate, which explained at best out of 17 covariates analyzed the high interpatient variability observed (9).

The relationship between the AUC of E7070 and the corresponding reduction in neutrophil and platelet counts has been explored within the context of the entire Phase I program (10).

Twenty-nine of the 37 patients in group A and 16 of 18 patients in group B were evaluable for response. One partial response was documented in a 68-year-old patient with endometrial adenocarcinoma previously treated with surgery and radiotherapy. The response detected in lymph node metastases was observed after the first cycle of E7070 at a dosage of 320 mg/m²/week and persisted until death during her fourth cycle of E7070 from an unrelated subdural hematoma. Her platelet count at the time of death was 164 × 10⁹/liter. A 46-year-old patient with locally metastatic melanoma to frontal cutaneous satellites and submandibular lymph node metastases who had previously received adjuvant IFN and chemoimmunotherapy for metastatic disease received a total of 21 cycles of E7070. During this period of treatment, his satellite metastases disappeared, and the metastatic lymph nodes shrank in size. A lymph node biopsy performed after the twenty-first cycle revealed necrotic material only, and treatment was discontinued. The submandibular lymph nodes remained unchanged in size for 1 year after discontinuation of E7070 before the patient undergoing lymph node dissection. The pathological specimen obtained revealed peripheral areas of viable but seemingly static melanoma in 5 of 24 lymph nodes examined.

Twelve patients achieved disease stabilization for a median of 5.3 months (range, 1.9–30.6+ months).

In this Phase I study of E7070 administered weekly on 4 consecutive weeks with each cycle being repeated at 6-week intervals, the MTD was 500 mg/m²/week, and the RD for further study was 400 mg/m²/week. These dose levels could not be safely increased in a preselected group of patients who were expected to tolerate higher doses of E7070. Reversible myelosuppression, particularly neutropenia and thrombocytopenia, were dose limiting. At the RD of 400 mg/m²/week dose, reduction was necessary in only 17% of patients in both groups. Additional treatment-related DLTs included stomatitis and hyperglycemia. Although intensive serum glucose monitoring was performed, hypoglycemia was not observed, and in contrast, one episode of hyperglycemia was noted in a patient with untreated diabetes mellitus. A single patient experienced rigors after the fourth dose of E7070. No explanation was found, and the rigors resolved the same day. The patient was not re-exposed to E7070. One patient died from septic shock complicated by multiorgan failure 2 days after receiving the second dose of E7070 on day 10. All other DLTs were reversible. The most common additional toxicities included anorexia, fatigue, diarrhea, fever, nausea, and vomiting, as well as taste perversion. They were all manageable with supportive therapies. No disturbances in hemodynamic function were identified, and no clinically significant cardiac arrythmias were observed. Although the majority of patients received a relatively small number of doses of E7070, this study did not find clear evidence of cumulative toxicity.

Four Phase I trials exploring different administration schedules of E7070 have now been completed and the toxicity profile of the compound has been similar in each. A comparison of these schedules does not suggest a clear advantage in terms of the dose intensity that can be safely delivered using any specific regimen (4, 5, 7). Although it is difficult to compare directly the efficacy and toxicity of E7070 when administered in a variety of different schedules, the nature of the DLTs observed has been consistent across the program. No clear evidence of schedule dependence was seen. The dose intensity of E7070 at MTD in the Phase I program is relatively consistent across schedules (267–333 mg/m²/week) with the exception of a lower level being tolerated after a continuous 5-day infusion of E7070 (217 mg/m²/week). This may partially reflect the limited number of dose levels explored within that study (7).

The pharmacokinetic profile of E7070 was nonlinear with a disproportional increase in exposure with increasing dose. This could be demonstrated with both a standard noncompartmental approach and a population pharmacokinetic model. Plasma concentration-time data were best described using a three-compartment model with nonlinear distribution to a peripheral compartment and two parallel pathways of elimination from the central compartment: a linear and a saturable pathway. In 1 patient, dose escalation was permitted, and the pharmacokinetics were analyzed additionally in cycles 14 and 18 (Fig. 2, A and B). The significantly increased AUCs at the two higher dosages of this patient clearly reflect and underline the nonlinear character of the pharmacokinetic behavior exhibited in the study population (Table 4, Fig. 3). Dose dependency of pharmacokinetic parameters was found in the same patient when stepwise dose escalated from 40 up to 320 mg/m²/week (Fig. 2, A and B). As E7070 displayed nonlinear pharmacokinetics at clinically relevant doses and the use of noncompartmental methods resulted in an inaccurate description of the complex concentration-time profile according to van Kesteren et al. (9), the more flexible population pharmacokinetic approach was applied. The model also provided accurate prediction of the plasma concentrations of the weekly times four schedule of this study. The fact that 1 patient when he had been escalated from initially 40–80 mg/m²/week (at cycle 9) and additionally to 160 mg/m²/week (at cycle 13) and additionally to 320 mg/m²/week (at cycle 14) experienced hematotoxicity but nevertheless tolerated additional 7 cycles without any dose delay when de-escalated to 160 mg/m²/week was in accordance with the elimination of E7070 described to be saturable and the presence of a linear elimination pathway preventing the occurrence of an unlimited accumulation of E7070 simultaneously. As a result of all population pharmacokinetic analyses of the four Phase I studies with E7070 performed (4, 5, 6, 7), it was shown that the body surface area-guided dosing is warranted for the safe administration of E7070 (9).

Myelosuppression after treatment with E7070 can be accurately predicted by the application of a pharmacodynamic model derived from the entire Phase I program (10). For the daily times five and the 120-h infusion, a 50% reduction in total count of WBCs, neutrophils, and platelets was reached at AUCs that were decreased to 52, 44, and 27% as compared with the AUCs with the 1-h and the weekly regimen, used in this study, for the corresponding variables. The authors concluded further that the potency of E7070 to produce hematological toxicity depends on the treatment regimen (10). This may explain the general low level of hematotoxicity observed in this study.

The peak concentrations of E7070 achieved at the RD of 400 mg/m²/week are in excess of the IC₅₀s required for cell death in >90% of human tumor cell lines tested (E7070 Investigator’s Brochure, version 6). As predicted from preclinical studies, tumor shrinkage and stabilization have been observed in a variety of different tumor types.

E7070 showed antitumor efficacy with one partial remission in a patient with endometrial cancer and disease stabilization in 12 patients. Notably, 1 patient with melanoma reached a stable disease lasting almost 45 months.

The RD of E7070 provides predictable and largely reversible toxicity and has been taken forward into Phase II studies with E7070 both as single agent and in combination with existing therapies.

We thank NOTOX B. V. for their contribution to the pharmacokinetic analysis of this study.