|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2002.11.061 on July 29 2002 © 2002 American Society for Clinical Oncology Phase I Study of Recombinant Human Endostatin in Patients With Advanced Solid TumorsByFrom the Departments of Thoracic and Head and Neck Medical Oncology, Biostatistics, Diagnostic Radiology, Pharmaceutical Sciences, Cancer Biology, Radiation Oncology, Surgical Oncology, and Gastrointestinal Medical Oncology, University of Texas M.D. Anderson Cancer Center, Houston, and University of Texas Houston Health Science Center, Houston, TX, and National Cancer Institute, Bethesda, MD. Address reprint requests to Roy S. Herbst, MD, PhD, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Blvd, Box 432, Houston, TX 77030.
PURPOSE: Endostatin, a 20-kd fragment of collagen XVIII, is a potent inhibitor of angiogenesis. We evaluated recombinant human endostatin (rh-Endo) in a phase I trial designed to assess safety, pharmacokinetics, and serum markers of angiogenesis in patients with solid tumors. PATIENTS AND METHODS: Twenty-six patients were enrolled onto a dose-finding trial of rh-Endo administered as an intravenous bolus over a 20-minute period once daily. Three patients each were treated at dose levels of 15, 30, 60, 120, 180, and 600 mg/m2/d, and seven patients were treated at 300 mg/m2/d. Treatment consisted of a minimum of two 28-day cycles. Evaluations included noninvasive imaging, pharmacokinetics, and serum biomarkers. RESULTS: Twenty-five patients were treated with rh-Endo. Treatment was well tolerated; there were no dose-limiting toxic effects. Bacteremia from frequent central line access was the most common problem. The pharmacokinetic disposition of rh-Endo was linear and best described using a two-compartmental open model. The overall mean half-life was 10.7 ± 4.1 hours. A dose of 300 mg/m2 achieved an area under the concentration-time curve associated with activity in preclinical models. In two patients, there was evidence of antitumor activity, but no responses were seen. Serum markers of angiogenic activity did not provide insight into rh-Endo’s activity. Serum antibodies were observed against both rh-Endo and the Pichia pastoris vector, but no allergic reactions were observed. CONCLUSION: rh-Endo was safe and well tolerated. rh-Endo pharmacokinetic profiles achieved area under the concentration-time curves associated with activity in preclinical models. Evidence of minor antitumor activity was observed and further studies are indicated.
TUMOR GROWTH is angiogenesis-dependent, and tumors  2 mm3 must recruit a new blood supply to remain metabolically active and expand in size.1-3 In addition, angiogenesis is vital to the overall process of metastasis, as micrometastases must establish new blood supplies for continued growth. Angiogenesis is therefore a prime target for anticancer drug development.4,5 Endostatin is an endogenous angiogenesis inhibitor that was first isolated from the supernatant of an in vitro culture of EOMA cells, a murine hemangioendothelioma cell line.6,7 Further characterization revealed that endostatin is a 20-kd C-terminal fragment of collagen XVIII. In in vitro cultures, human and murine endostatin specifically inhibit the proliferation and migration of capillary endothelial cells and can induce apoptosis of proliferating endothelial cells. However, no direct effect on the growth of numerous tumor cell lines has been seen.6 In several different assays in vivo, recombinant human endostatin (rh-Endo) and murine endostatin are potent inhibitors of angiogenesis. Endostatin was subsequently shown to inhibit, in a potent and dose-dependent manner, the growth of a wide variety of human and murine primary and metastatic tumors growing in mice.6,7 Prolonged therapy with high doses of endostatin induced a virtually complete blockade of tumor angiogenesis and caused established tumors to regress to microscopic lesions.7 Examination of these dormant lesions revealed decreased angiogenesis with little or no change in the rate of tumor cell proliferation but significantly increased tumor cell apoptosis as compared with untreated controls. No resistance to therapy or toxic effects was observed in mice even after prolonged therapy with endostatin.7,8 These striking results in experimental mouse models provided the impetus to initiate clinical trials of endostatin in patients with cancer. The gene encoding the 183 amino acids of human endostatin9 was cloned, and rh-Endo was subsequently expressed under the strongly inducible AOXI promoter in Pichia pastoris. Preclinical pharmacology and toxicology studies indicated that when weight-adjusted doses of rh-Endo were administered to monkeys, blood levels greater than those occurring in mice were achieved, and rh-Endo was biologically active and nontoxic.10 Further studies showed that rh-Endo had all the characteristics of native endostatin and effectively inhibited the growth of primary tumors and pulmonary metastases in a dose-dependent manner. These same investigators also showed that deletion of two of the four zinc ligands of rh-Endo did not affect its activity.10 Extensive animal pharmacologic studies were performed with rh-Endo at the effective antitumor doses beginning at 1.5 mg/kg. Depending on the dose administered, the area under the concentration-time curve (AUC) ranged from 16 to 700 ng/mL·h, and the maximum concentration (Cmax) ranged from 161 to 4,582 ng/mL. Thus, the preclinical data suggested that rh-Endo was active, was pharmacologically and toxicologically well behaved, and could be produced using recombinant technology in quantities sufficient for human use.10 Based upon these considerations, a phase I study of rh-Endo sponsored by the National Cancer Institute was initiated at the University of Texas M.D. Anderson Cancer Center. The primary clinical end points of this phase I study were to assess safety and tolerability and to measure the pharmacokinetic profile and interpatient pharmacologic variability of rh-Endo in patients with incurable solid tumors. The secondary end points of this trial were to evaluate tumor response rates, to measure the effects of rh-Endo on circulating proangiogenic factor levels, and to establish a recommended phase II dose of rh-Endo for future clinical studies.
Patient Selection Patients with advanced solid malignancies for which no standard curative therapy is known were eligible for this study. All patients were required to have at least one site of measurable and previously unirradiated disease. Other eligibility criteria included the following: age 18 years, an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1, the absence of serious medical or psychiatric disorders, adequate renal function (serum creatinine < 1.5 mg/dL or a calculated creatinine clearance > 60 mL/min), adequate bone marrow function (WBC count > 3,000/mm3, absolute neutrophil count [ANC] > 1,500 mm3, platelet count > 100,000/mm3, and hemoglobin > 10 g/dL), and adequate hepatic function (total bilirubin value < 1.5 times the upper limit of normal [ULN], ALT level < 2.0 times ULN, AST < 2.0 times ULN, and prothrombin time/partial thromboplastin time < 1.5 times normal). A tissue biopsy was required at study entry and again at the end of the second treatment cycle, and patients with easily accessible tumors were preferred for enrollment. Patients were informed about the experimental nature of this program and, according to institutional and federal guidelines, signed an approved informed consent form before investigational diagnostic studies or therapy with rh-Endo began. Patients were excluded if they were pregnant or breast feeding, had a history of brain metastases or a primary brain tumor, or had a history of hemorrhagic diathesis. All women of childbearing potential were required to have a negative pregnancy test. All sexually active patients of childbearing potential were required to practice adequate contraception for the entire treatment period. Patients could not have undergone minor surgery (eg, central venous catheter placement) within 24 hours of treatment with rh-Endo or any major surgery (eg, laparotomy, thoracotomy, or craniotomy) within 4 weeks of enrollment. Patients who required surgical procedures while on trial had their rh-Endo stopped 3 days before the procedure. Patients were advised not to use herbal remedies or other over-the-counter biologic agents. Heparin products were not used because of the binding of rh-Endo to heparin and the potential effects of heparin on angiogenesis. There was no limit on the amount of prior chemotherapy allowed for patients with an ECOG performance status of 0; however, patients with an ECOG performance status of 1 could not have received more than three prior chemotherapy regimens. Prior adjuvant treatment for nonmetastatic disease or as part of a concomitant chemoradiation protocol was allowed and was not included in the three-regimen limit. If the patient had received prior chemotherapy, the following waiting periods were required between the final dose of that prior therapy and the initiation of treatment with rh-Endo: nitrosoureas or mitomycin, 6 weeks; any investigational agent, 30 days; any other cytotoxic or cytostatic agent, 3 weeks. A negative magnetic resonance imaging (MRI) or computed tomography (CT) scan of the brain was required of all patients before entry. Patients could have received prior radiation therapy to sites that were not assessed for response, but such patients were required to wait 3 weeks after the completion of treatment before enrollment onto this trial. To facilitate repeated vascular access, all patients were required to have a central venous catheter inserted.
Drug Administration and Study Design
Patients were assigned to a predetermined dose level as shown in Table 1. The first cohort of patients was treated at a dose of 15 mg/m2/d. Subsequent cohorts of three patients each were treated at 30, 60, 120, 180, and 300 mg/m2/d. Decisions to escalate to the next higher dose level were deferred until at least three patients at each level had been observed for at least 4 weeks. If no dose-limiting toxic effects had occurred, three patients were accrued for treatment at the next dose level. At the 300-mg/m2/d level, death of a patient due to disease progression prompted expansion of this level to seven patients to better define the pharmacokinetic and safety profile. Subsequently, three patients were treated at 600 mg/m2/d. Dose escalation was halted at that point because of trial design specification and drug availability concerns expressed earlier. In the absence of significant (< grade 2) toxic effects, patients who had received at least 8 weeks of therapy could be treated at the next highest dose level when three rh-Endo–naïve patients had completed one course of therapy at the higher level without toxic effects. Throughout the study, pharmacokinetic data were obtained from and evaluated for all patients. To ensure patient safety, discussions were held with the NCI before patients were treated at each new dose level.
After each treatment cycle, the following studies were performed: routine laboratory evaluation (including complete blood count, evaluation of liver and renal function, and assessment of beta-human chorionic gonadotropin in women of reproductive potential), routine radiographic imaging, functional imaging, and serum proangiogenic biomarker studies. The results of the serial noninvasive imaging studies will be reported separately. Because preclinical findings showed that tumors often initially grew during endostatin treatment before response became evident, this study allowed patients to continue on the study with up to 100% bidimensional tumor growth in the absence of new symptoms.7 However, for the purpose of trial reporting, a 25% growth increase was used to define progression.
Pharmacokinetic Assessment Before analysis, blood samples were allowed to clot and were then centrifuged at 1,500 rpm for 10 minutes at 4°C. The resulting serum was divided into at least three 0.5-mL aliquots and frozen at -80°C until time for analysis. To preserve protein integrity, all specimens were thawed only once.8 The rh-Endo serum concentration was determined using an enzyme immunoassay (EIA) kit provided by Entremed. The Accucyte endostatin EIA uses immobilized goat antirabbit antibodies to bind a purified rabbit antihuman endostatin polyclonal antibody. Free endostatin in the sample was incubated with the immobilized rabbit antihuman endostatin polyclonal antibody along with biotinylated human endostatin, which was added in amounts that were known to generate a competitive binding reaction. Thus, as the concentration of rh-Endo in the serum sample increased, the amount of biotinylated endostatin bound by the polyclonal antibody decreased. After the addition of streptavidin-conjugated alkaline phosphatase followed by the addition of a chromogenic reagent solution, the amount of biotinylated endostatin was quantified colorimetrically.10 The Accucyte EIA was validated at M.D. Anderson Cancer Center by confirmation of results from quality control samples containing known amounts of rh-Endo. Data are reported as the mean ± 1 SD at each dose level. Pharmacokinetic parameters were determined by compartmental modeling of the serum concentration-time data obtained from each patient using maximum likelihood estimation. The final models were selected using Akiake’s information criteria, visual inspection, and statistical estimation of goodness of fit. Parameters such as the volume of distribution of the central compartment, the elimination rate constant, and microconstants were estimated, while steady-state volume of distribution, half-lives, and clearance were calculated from the primary parameters. AUC and Cmax were determined from the estimated parameters. Pharmacokinetic modeling was performed using ADAPT II Pharmacokinetic and Pharmacodynamic Systems Analysis Software Version 4.0 (Biomedical Simulations Resource, University of Southern California, Los Angeles, CA).11
Serum Assays of Proangiogenic Factors
Immunologic Assays
Statistical Methods For the serum biomarkers, we used exploratory graphical analyses to examine the trajectories of values for each patient over time. We also plotted early time changes (baseline to 1 month, 1 month to 2 months, and baseline to 2 months) against dose. Association between relative changes in serum values over time and dose were assessed with rank correlation analyses.
Patient Characteristics and Treatments Twenty-six patients were enrolled onto this trial. Patient characteristics are shown in Table 1. Toxic effects and responses to therapy in 25 patients were assessable; one patient was not treated because symptomatic disease progression began before therapy started. All patients were treated according to a protocol-specified escalation schedule with planned dose levels of 15, 30, 60, 120, 180, 300, and 600 mg/m2/d (Table 1). Each cohort consisted of three patients, with the exception of the 300-mg/m2 cohort, which enrolled seven patients. Five patients received one dose and the next higher dose, on the basis of the criteria described in the Patients and Methods (Table 1). The median age of the 25 treated patients was 56 years (range, 29 to 78 years). There were 17 men and eight women; all had a median performance status of 1. Different malignancies were treated: eight patients (31%) had melanoma, five (19%) had sarcoma, three (12%) each had breast or lung cancer, two (8%) each had colorectal, head and neck, or renal cancer, and one (4%) had thyroid cancer. This was a heavily pretreated population. Only three patients (12%) had no prior cytotoxic therapy. Twelve patients (46%) had one to two prior regimens, seven (27%) had three to five, and four (15%) had six to 10 prior cytotoxic chemotherapy regimens.
Toxic Effects
Hematologic toxic effects were rare; grade 3 anemia and grade 3 lymphocytopenia each occurred once. One patient had both grade 3 anemia and grade 2 thrombocytopenia during treatment. This patient had indolent malignant melanoma and had been heavily exposed to chemotherapy with alkylating agents. Bone marrow analysis documented an evolving myelodysplastic syndrome. After discontinuation of rh-Endo, the patient’s platelet counts did not return to normal values.
Pharmacokinetics of rh-Endo When administered intravenously, rh-Endo demonstrated a linear pharmacokinetic profile across all dose levels. A representative serum concentration-time profile from a patient treated with 300 mg/m2 is shown in Fig 2. Using our sampling schema, the disposition of rh-Endo was best described by a two-compartment open model. Clearance, Cmax, and AUC relative to each rh-Endo dose are outlined in Table 3. The mean t1/2 beta and clearance were 10.7 ± 4.1 hours and 408 ± 127 mL/min/m2, respectively, and were similar across all dose levels. The mean steady-state volume of distribution was 50.4 ± 25.2 L/m2. The trough concentration increased with each increase in dose level of rh-Endo, and little accumulation was observed with repeated doses of rh-Endo. Trough concentrations reached steady state after a week of therapy (Table 3). Limited repeat sampling at day 28 due to noncompliance made determination of intraindividual variability impossible.
In tumor-bearing mice, activity against Lewis lung carcinoma was documented after subcutaneous injection of 50 mg/kg every 12 hours.10 Pharmacokinetic studies in tumor-free mice after intravenous injection of rh-Endo 50 mg/kg produced a Cmax of 59.4 µg/mL and an AUC of 660 µg/mL·min.10 Thus, an AUC of 600 to 700 µg/mL·min represented a target for the human phase I studies. The 300-mg/m2/d and 600-mg/m2/d doses achieved mean exposures at or in excess of this target AUC (Fig 3A). There was no relationship between body size (body surface area) and rh-Endo plasma clearance (Fig 3B).
Time on Study/Antitumor Activity Patients remained in this phase I study for a median period of 69 days; the range was 14 to 426 days (Fig 4). The median time to tumor progression was 45 days using standard response criteria, whereby 25% bidimensional growth defined progressive disease. In two of 25 patients, some evidence of an anticancer effect was demonstrated, although there were no formal partial or complete responses seen. Patient no. 2 had metastatic malignant melanoma with multiple dermal metastases. He had previously experienced progressive disease despite 10 prior chemotherapy regimens (Table 2). The patient’s dermal metastases initially grew during the first 8 weeks of therapy, then stabilized during subsequent courses of rh-Endo (Fig 5). With gradual escalation of the rh-Endo dose from 15 mg/m2/d to 300/mg/m2/d (Fig 5), a single lesion over the left mandible showed some improvement, and no new lesions developed during the entire period of therapy (426 days, 14 courses). Prolonged stabilization of disease translated into a much improved quality of life for this patient, and he continued to work while he participated in the trial. The patient was removed from the trial when a posterior occipital lesion involving the scalp and underlying calvarium progressed, requiring neurosurgical intervention. A patient with recurrent synovial cell sarcoma (Fig 6) who had undergone surgery, radiotherapy, and three regimens of chemotherapy had a mixed tumor response after two courses of therapy. A submandibular lesion regressed by 60% and was associated with marked symptomatic improvement in pain and improved head mobility during the treatment period (Fig 6B). However, an infratemporal lesion progressed by more than 100% (Fig 6A). This patient was taken off the study after 143 days of treatment with rh-Endo because the infratemporal lesion continued to grow.
Serum Markers Assays were conducted to assess the effect of rh-Endo on the levels of four different serum proteins thought to be important mediators of angiogenesis and metastasis. The levels were measured before the study and after treatment at 28-day intervals. The data are reported in Fig 7. Shown are serum VEGF/vascular permeability factor (adjusted for platelet count), bFGF, VCAM, and E-selectin levels as described by Byrne et al.12 There was considerable variation in the baseline levels of all four factors. Furthermore, after initiation of therapy with rh-Endo, there were no consistent changes in the levels of any of these proteins. The absence of a measurable effect on VEGF, bFGF, VCAM, and E-selectin also included the two patients who experienced minor anticancer effects.
Immunogenicity of rh-Endo Studies to assess the immunogenicity of rh-Endo were performed on a subset of treated patients. Sera from 13 patients were assessed for IgG (Fig 8A) and IgM (Fig 8B) antibodies to rh-Endo. No patient had preexisting IgG immunoreactivity to endogenous endostatin. After the initial course of treatment (28 days), nine (69%) of the 13 patients were positive for IgG immunoreactivity to the rh-Endo formulation used in the trial. Titers ranged from 1,000 to 100,000. The median titer was 5,000. After two cycles of treatment, eight (73%) of 11 patients had results that were positive for rh-Endo protein IgG immunoreactivity, with titers ranging from 6,400 to 12,800. The data were similar for IgM immunoreactivity (Fig 8B). Taken together, these data strongly suggest that immunoreactivity to the rh-Endo formulation used in this trial is common.
To assess whether the antibodies generated against rh-Endo were specifically directed against the recombinant endostatin protein, detailed studies were conducted on the patient with metastatic melanoma who remained on therapy for 14+ months. After 400 days of therapy, the patient’s titer of IgG immunoreactivity against rh-Endo was 32,000. Using identical methodology, IgG immunoreactivity to a yeast protein extract from P pastoris was assessed. The titer was found to be 12,800. These data suggest that the immunoreactivity in this patient was directed to both rh-Endo and an unknown Pichia protein. Additional studies will be needed to clarify the immunogenicity of rh-Endo formulations and to determine whether there is any clinical or pharmacologic impact to antibody formation.
This study was primarily designed to assess the safety and pharmacokinetic profile of human recombinant endostatin. We found that rh-Endo was safe at doses ranging from 15 to 600 mg/m2 /d and that people receiving those doses did not suffer intolerable side effects. A minority of patients experienced grade 3 toxicities (most often associated with the central venous line) that were not directly attributed to rh-Endo. Pharmacologically, rh-Endo was well behaved, with linear pharmacokinetics within the dose range studied. On the basis of preclinical pharmacodynamic studies performed with the Lewis lung carcinoma model, a target AUC of 600 to 700 µg/mL·min was established.10 In this trial, daily intravenous doses of 300 to 600 mg/m2/d achieved a similar level of exposure to rh-Endo. There was evidence of minor antitumor activity in two patients, who each had regression of at least one metastatic lesion during treatment. Serum assays of VEGF, bFGF, VCAM, and E-selectin demonstrated substantial interpatient heterogeneity but did not show any consistent change during therapy with rh-Endo, although this was not entirely unexpected given that little preclinical evidence currently exists to support this effect. The current formulation of rh-Endo produced from the yeast P pastoris was immunogenic, with antibodies to the current formulation of rh-Endo identified in all 13 patients studied. Preliminary studies suggest that some of the antibodies may be directed at yeast proteins that are minor contaminants of the current rh-Endo formulation. The functional significance of these antibodies is unknown, but they did not appear to contribute to any clinically relevant allergic reactions at the rh-Endo doses studied. Given the theoretical selectivity of angiogenesis inhibitors for tumor-related endothelial cells, it has generally been assumed that these agents will be nontoxic in clinical trials. However, this has not been demonstrated in all of the clinical studies reported to date. For example, early clinical trials with the anti-VEGF monoclonal antibody in non–small-cell lung cancer have been complicated by episodes of pulmonary hemorrhage,13 and the small molecule receptor tyrosine kinase (VEGF receptor 2) inhibitor SU5416 has produced more than the expected number of thrombotic events in early trials.14 It is tempting to speculate that these agents may have had an effect on normal vasculature, since VEGF/vascular permeability factor has been shown to be important in maintaining endothelial integrity.15 TNP-470, one of the earliest antiangiogenic agents studied, can produce neurocognitive side effects at high doses16,17 and is an abortive agent. Interestingly, to date, most angiogenesis inhibitors have not inhibited wound healing. The reasons may include the inhibitors’ relative specificity for tumor-related endothelial cells, the limited exposure times during clinical testing, and the selection of patients who are not likely to be challenged with major surgical procedures. In this trial, at doses from 15 mg/m2 to 600 mg/m2, endostatin had an excellent safety profile, which suggests that it can be easily combined with additional therapies. The pharmacokinetic analyses conducted during this trial successfully demonstrated that doses of rh-Endo that approached or exceeded peak serum concentration and AUCs associated with anticancer activity in preclinical models can be safely delivered in the clinic. Preclinical studies with Escherichia coli-derived rh-Endo and the P pastoris product used in this study have suggested an AUC target concentration of 600 to 700 µg/mL·min for clinical activity. In our study, the rh-Endo doses of 300 to 600 mg/m2/d achieved this targeted level. No relationship was observed between body surface area and endostatin clearance, and little intraindividual pharmacokinetic variability was seen in the limited subset of patients with repeat samples. However, validating this exposure as an appropriate target for future clinical development of rh-Endo will require additional study, including phase II studies with selective doses in more homogeneous populations. This fact is supported by the observation that anticancer activity was seen in a patient whose therapy began at the lowest dose level (patient no. 2, Fig 5). This patient’s dose was escalated four times while he was in the study, and he ultimately received a dose of 300 mg/m2 (Table 1). Interestingly, baseline levels of endogenous endostatin were detected in all patients and had to be accounted for in the pharmacokinetic analyses. In 25 healthy volunteers, the mean circulating endostatin concentration from random sampling was 30.9 ± 10.7 ng/mL. A correlation between low endostatin concentration and tumor progression has been suggested by some investigators.18-20 More recently, a higher incidence of prostate cancer has been described in patients who have a polymorphism in the endostatin gene. This group has speculated that this polymorphism (N104) may impair the functioning of endostatin.21 Other factors may complicate the conclusion that an AUC of 600 to 700 µg/mL·min is an appropriate surrogate for rh-Endo exposure that would be expected to demonstrate activity in phase II trials. For example, the ELISA used in this trial measures total endostatin protein but does not demonstrate active angiogenic inhibition. Proteolytic degradation of endostatin could certainly result in early inactivation yet not be reflected by the ELISA values. In this context, we have attempted to demonstrate antiendothelial cell apoptotic activity in sera from the patients on this study without success.22 The reasons for this are potentially complex, but it would appear that the ex vivo apoptosis assay used may not have been sensitive enough to demonstrate a biologic effect. Appropriate surrogate measures of antiangiogenic activity after administration of rh-Endo should be developed during future clinical trials. The linear pharmacokinetic behavior of rh-Endo suggests that approaches other than daily drug administration should be studied. Indeed, the daily intravenous schedule was very inconvenient for the patients. Switching to a continuous daily infusion of rh-Endo might be one way to circumvent this problem. Continuous exposure to endostatin is supported by a number of observations. In the original preclinical studies showing tumor regression, endostatin was administered using a sustained-release preparation that had more activity than did the soluble protein.6 More recently, Drixler et al23 found that the antitumor and antiangiogenic activities of the related, naturally occurring antiangiogenic peptide angiostatin were improved when angiostatin was administered by continuous infusion instead of bolus administration. On the basis of these studies, Kisker et al24 used the Lewis lung mouse model to show that, with equal doses, endostatin is more active with continuous exposure. Current clinical trials at the M.D. Anderson Cancer Center and at several other centers are now exploring this approach to rh-Endo administration. In this initial phase I trial, no complete or partial responses were documented. The expectations for rh-Endo were extraordinarily high to produce response, given the excellent preclinical data previously reported.7,25 The minor anticancer activity observed in this trial supports evolving laboratory data indicating that the clinical effects of antiangiogenic agents are likely to be complex.26,27 Part of the difficulty in translating preclinical efficacy directly to the clinical setting is the lack of predictive preclinical models. For example, animal models of angiogenesis are often conducted on tumors recently implanted in nude mice. In the human situation, tumors could have been present for months to years and have a more established vasculature that would be less sensitive to antiangiogenic therapy. It is also very possible that endostatin and other angiogenesis inhibitors must be given for longer periods of time to achieve an optimal effect or must be used in more minimal disease settings. This may be important because in preclinical studies with angiogenesis inhibitors, some tumors actually progressed before eventually responding to therapy.28 Additionally, data are beginning to emerge that suggest a heterogeneity of the tumor vasculature in different organs.29-31 Different tumor clones in unique microenvironments might respond differently to a single angiogenesis inhibitor. In addition, we cannot fully exclude the possibility that antibodies formed against rh-Endo might limit its efficacy. Further study is warranted. The traditional end point of using toxicity as a surrogate for biologic activity may not be relevant in assessing biologic agents, which may have a much broader therapeutic index.32 If toxic effects are mediated by a mechanism distinct from that of antitumor action, increasing doses to toxic levels would not be necessary to achieve efficacy. This is especially important for an apparently nontoxic agent such as rh-Endo, for which the ultimate goal is to identify an optimal biologic dose. A critical issue in the early development of antiangiogenic agents has been the parallel development of appropriate surrogate end points that correlate with antiangiogenic activity and can predict clinical efficacy. Generally, surrogate end points are important for determining efficacy of compounds that may not induce frank tumor regression. The assessment of angiogenic proteins in serum was not informative in this study. This is consistent with the results from studies of other antiangiogenic agents and suggests the need for identifying and validating correlative biologic markers in preclinical models. Possible gauges of activity could include measurement of the direct and indirect effects of the antiangiogenic agent on specific molecular targets.32 Although endostatin has been shown to interact with integrins33 and to bind to an endothelial isoform of tropomyosin with effects on cell migration,34,35 a definitive receptor and mechanism of action has still not been defined. We therefore designed this trial to include serial measurements of tumor blood flow, tumor metabolism, and the effects of rh-Endo on endothelial cell and tumor cell viability. These data will be reported separately.36 The results of this phase I study demonstrate that endostatin is safe and has a linear pharmacokinetic profile when administered at doses up to 600 mg/m2/d. Future studies will be required to examine the single-agent activity of endostatin administered as an intravenous bolus or continuous infusion or in combination with chemotherapy. Additionally, there exists the possibility of combining endostatin with other antiangiogenic agents. Preclinical data support the validity of all of these options (O’Reilly et al, manuscript in preparation).
Supported by grant no. U01 CA 62461 from the United States Public Health Service, Cancer Center Support grant no. CA 16672, and the Golfers Against Cancer. We thank Diane Gravel, Karen Terry, and Mercedes Guerra for nursing, Drs Merrick Ross, Barry Feig, and Paul Mansfield for biopsies, and Anh Lee, John Nguyen, Sonya Dalton, and Bich Tran for assistance.
Presented in part at the Thirty-Seventy Annual Meeting of the American Society of Clinical Oncology, San Francisco, CA, May 12-15, 2001. This article was published ahead of print at www.jco.org.
1. Folkman J: What is the evidence that tumors are angiogenesis dependent? J Nat Cancer Inst 82: 4-6, 1990 2. Folkman J: Tumor angiogenesis, in Mendelsohn J, Howley RM, Israel MA, et al (eds): The Molecular Basis of Cancer. Philadelphia, PA, WB Saunders, 1995, pp 206-232 3. Folkman J: Tumor angiogenesis. Adv Cancer Res 19: 331-358, 1974[Medline] 4. Fidler IJ: Cancer biology: Invasion and metastasis, in Abeloff MD, Armitage JO, Lichter AS, et al (eds): Clinical Oncology, ed 2. New York, NY, Churchill Livingston, 2000, pp 29-53 5. Fidler IJ, Ellis L: The implications of angiogenesis for the biology and therapy of cancer metastasis. Cell 79: 185-188, 1994[CrossRef][Medline] 6. O’Reilly MS, Boehm T, Shing Y, et al: Endostatin: An endogenous inhibitor of angiogenesis and tumor growth. Cell 88: 277-285, 1997[CrossRef][Medline] 7. Boehm T, Folkman J, Browder T, et al: Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance. Nature 390: 404-407, 1997[CrossRef][Medline] 8. Sim BK, MacDonald NJ, Gubish ER: Angiostatin and endostatin: Endogenous inhibitors of tumor growth. Cancer Metastasis Rev 19: 181-190, 2000[CrossRef][Medline] 9. Oh SP, Warman ML, Seldin MF, et al: Cloning of cDNA and genomic DNA encoding human type XVIII collagen and localization of the alpha 1(XVIII) collagen gene to mouse chromosome 10 and human chromosome 21. Genomics 19: 494-499, 1994[CrossRef][Medline] 10. Sim BKL, Fogler WE, Zhou XH, et al: Zinc ligand-disrupted recombinant human endostatin: Potent inhibition of tumor growth, safety and pharmacokinetic profile. Angiogenesis 3: 41-51, 1999[CrossRef][Medline] 11. D’Argenio DZ, Schumitzky A: ADAPT II, User’s Guide: Pharmacokinetic/Pharmacodynamic Systems Analysis Software. Los Angeles, CA, Biomedical Simulations Resource, University of Southern California, 1997
12. Byrne GJ, Ghellal A, Iddon J, et al: Serum soluble vascular cell adhesion molecule-1: Role as a surrogate marker of angiogenesis. J Natl Cancer Inst 92: 1329-1936, 2000 13. DeVore R, Fehrenbacher L, Herbst R, et al: A randomized phase II trial comparing rhuMAb VEGF (recombinant humanized monoclonal antibody to vascular endothelial cell growth factor) plus carboplatin/paclitaxel chemotherapy (CRx) to CRx alone in patients with stage IIIb/IV NSCLC. Proc Am Soc Clin Oncol 19: 485a, 2000 (abstr 1896)
14. Kuenen BC, Rosen L, Smith EF, et al: Dose-finding and pharmacokinetic study of cisplatin, gemcitabine, and SU5416 in patients with solid tumors. J Clin Oncol 20: 1657-1667, 2002 15. O’Reilly MS: Antiangiogenesis: Basic principles, in Rosenberg SA (ed): Principles and Practice of the Biologic Therapy of Cancer, ed 3. Philadelphia, PA, Lippincott Williams & Wilkins, 2000, pp 827-843 16. Herbst RS, Madden TL, Tran HT, et al: Safety and pharmacokinetic effects of TNP-470, an angiogenesis inhibitor, combined with paclitaxel in patients with solid tumors: Evidence for activity in non–small-cell lung cancer. J Clin Oncol (in press)
17. Logothetis CJ, Wu KK, Finn LD, et al: Phase I trial of the angiogenesis inhibitor TNP-470 for progressive androgen-independent prostate cancer. Clin Cancer Res 7: 1198-1203, 2001
18. Yamagata M, Shiratori Y, Dan Y, et al: Serum endostatin levels in patients with hepatocellular carcinoma. Ann Oncol 11: 761-762, 2000 19. Feldman AL, Pak H, Yang JC, et al: Serum endostatin levels are elevated in patients with soft tissue sarcoma. Cancer 91: 1525-1529, 2001[CrossRef][Medline]
20. Feldman AL, Tamarkin L, Paciotti GF, et al: Serum endostatin levels are elevated and correlate with serum vascular endothelial growth factor levels in patients with stage IV clear cell renal cancer. Clin Cancer Res 6: 4628-4634, 2000
21. Iughetti P, Suzuki O, Godoi PHC, et al: A polymorphism in endostatin, an angiogenesis inhibitor, predisposes for the development of prostatic adenocarcinoma. Cancer Res 61: 7375-7378, 2001 22. Roach JS, Baker CH, Tran HT, et al: Induction of apoptosis of human endothelial cells by recombinant endostatin: Development of surrogate assay for clinical studies. Proc Am Assoc Cancer Res 42: 206, 2002 (abstr 1105)
23. Drixler TA, Rinkes IH, Ritchie ED, et al: Continuous administration of angiostatin inhibits accelerated growth of colorectal liver metastases after partial hepatectomy. Cancer Res 60: 1761-1765, 2000
24. Kisker O, Becker CM, Prox D, et al: Continuous administration of endostatin by intraperitoneally implanted osmotic pump improves the efficacy and potency of therapy in a mouse xenograft tumor model. Cancer Res 61: 7669-7674, 2001
25. Pentz RD, Flamm AL, Sugarman J, et al: Study of the media’s influence on prospective research participants’ understanding of and motivations for participation in a high-profile phase I trial. J Clin Oncol 20: 3785-3791, 2002
26. Baish JW, Jain RK: Fractals and cancer. Cancer Res 60: 3683-3688, 2000 27. Brown EB, Campbell RB, Tsuzuki Y, et al: In vivo measurement of gene expression, angiogenesis and physiological function in tumors using multiphoton laser scanning microscopy. Nat Med 7: 864-868, 2001[CrossRef][Medline] 28. O’Reilly MS, Holmgren L, Chen C, et al: Angiostatin induces and sustains dormancy of human primary tumors in mice. Nat Med 2: 689-692, 1996[CrossRef][Medline]
29. Arap W, Pasqualini R, Ruoslahti E: Cancer treatment by targeted drug delivery to tumor vasculature in a mouse model. Science 279: 377-380, 1998 30. Bicknell R: Vascular targeting and the inhibition of angiogenesis. Ann Oncol 5: 45-50, 1994
31. Korn EL, Arbuck SG, Pluda JM, et al: Clinical trial designs for cytostatic agents: Are new approaches needed? J Clin Oncol 19: 265-272, 2001
32. Eisenhauer EA: Phase I and II trials of novel anti-cancer agents: Endpoints, efficacy and existentialism—The Michel Clavel Lecture, held at the 10th NCI-EORTC Conference on New Drugs in Cancer Therapy, Amsterdam, 16-19 June 1998. Ann Oncol 9: 1047-1052, 1998
33. Rehn M, Veikkola T, Kukk-Vladre E, et al: Interaction of endostatin with integrins associated with angiogenesis. Proc Natl Acad Sci U S A 98: 1024-1029, 2001
34. MacDonald NJ, Shivers WY, Narum DL, et al: Endostatin binds tropomyosin: A potential modulator of the antitumor activity of endostatin. J Biol Chem 276: 25190-25196, 2001
35. Ackley BD, Crew JR, Elamaa H, et al: The NC1/endostatin domain of Caenorhabditis elegans type XVIII collagen affects cell migration and axon guidance. J Cell Biol 152: 1219-1232, 2001
36. Herbst RS, Mullani NA, Davis DW, et al: Development of biologic markers of response and assessment of antiangiogenic activity in a clinical trial of human recombinant endostatin. J Clin Oncol 20: 3804-3814, 2002 Submitted November 13, 2001; accepted July 2, 2002. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
