|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology
Phase II Trial of the Antiangiogenic Agent Thalidomide in Patients With Recurrent High-Grade GliomasFrom the Center for Neuro-Oncology, Dana-Farber Cancer Institute, and Departments of Neurology, Surgery, and Radiation Oncology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA; Department of Neuro-Oncology, M.D. Anderson Cancer Center, and National Central Nervous System Consortium, Houston, TX; and Cancer Treatment and Evaluation Program, Investigational Drug Branch, Division of Cancer Therapy and Diagnosis, National Cancer Institute, Bethesda MD. Address reprint requests to Howard A. Fine, MD, Neuro-Oncology Branch, National Cancer Institute, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Building 10, Rm 12W253, 10 Center Dr, Bethesda, MD 20892-2440; email hf32p{at}nih.gov ABSTRACT PURPOSE: Little progress has been made in the treatment of adult high-grade gliomas over the last two decades, thus necessitating a search for novel therapeutic strategies. Malignant gliomas are vascular or angiogenic tumors, which leads to the supposition that angiogenesis inhibition may represent a potentially promising strategy in the treatment of these tumors. We present the results of a phase II trial of thalidomide, a putative inhibitor of angiogenesis, in the treatment of adults with previously irradiated, recurrent high-grade gliomas. PATIENTS AND METHODS: Patients with a histologic diagnosis of anaplastic mixed glioma, anaplastic astrocytoma, or glioblastoma multiforme who had radiographic demonstration of tumor progression after standard external-beam radiotherapy with or without chemotherapy were eligible. Patients were initially treated with thalidomide 800 mg/d with increases in dose by 200 mg/d every 2 weeks until a final daily dose of 1,200 mg was achieved. Patients were evaluated every 8 weeks for response by both clinical and radiographic criteria. RESULTS: A total of 39 patients were accrued, with 36 patients being assessable for both toxicity and response. Thalidomide was well tolerated, with constipation and sedation being the major toxicities. One patient developed a grade 2 peripheral neuropathy after treatment with thalidomide for nearly a year. There were two objective radiographic partial responses (6%), two minor responses (6%), and 12 patients with stable disease (33%). Eight patients were alive more than 1 year after starting thalidomide, although almost all with tumor progression. Changes in serum levels of basic fibroblastic growth factor (bFGF) were correlated with time to tumor progression and overall survival. CONCLUSION: Thalidomide is a generally well-tolerated drug that may have antitumor activity in a minority of patients with recurrent high-grade gliomas. Future studies will better define the usefulness of thalidomide in newly diagnosed patients with malignant gliomas and in combination with radiotherapy and chemotherapy. Additionally, studies will be needed to confirm the potential utility of changes in serum bFGF as a marker of antiangiogenic activity and/or glioma growth. THE REALIZATION THAT some high-grade or malignant primary brain tumors such as CNS lymphomas, primitive neuroectodermal tumors, and anaplastic oligodendrogliomas are often sensitive to standard cytotoxic agents has led to advances in the treatment of these tumors.1-3 Nevertheless, the treatment of patients with high-grade astrocytomas, the most common primary brain tumors in adults, has seen few improvements since the demonstration more than two decades ago that radiation therapy prolongs survival.4 Clearly, new strategies aimed at novel therapeutic targets are needed. More than 20 years ago, Folkman5 proposed the hypothesis that solid tumor growth was dependent on the development of tumor-associated blood vessels, a process called angiogenesis. Numerous studies of experimental and human tumors have confirmed the central role of angiogenesis in solid tumor progression, including high-grade gliomas (see review in6-9 ). Consistent with the importance of angiogenesis in the biology of astrocytomas is the recognition of endothelial proliferation as a marker for high-grade or aggressive glioma histology in several different grading classifications and the demonstration that the degree of microvascularity, as assessed by endothelial cell/capillary density, correlates with the biologic aggressiveness of astrocytic tumors.10 Additionally, a growing body of cellular and molecular data support the central role of angiogenesis in the biology of glioma progression. For example, it has been demonstrated that the degree of basic fibroblastic growth factor (bFGF) expression within an astrocytoma is positively correlated with the grade of the tumor.11,12 It has also been demonstrated that malignant glioma cells overexpress vascular endothelial growth factor (VEGF), a highly specific vascular endothelium mitogen.8,9 This overexpression is most dramatic in the tumor in areas adjacent to necrosis, which is consistent with the known hypoxia inducibility of the VEGF promoter. A natural corollary to the hypothesis that solid tumors, like gliomas, are dependent on angiogenesis is the idea that inhibition of angiogenesis could prove to be a novel therapeutic approach to controlling tumor growth. Such an approach is theoretically attractive because physiologic angiogenesis in the adult is largely limited to wound healing and the female reproductive tract during the menstrual cycle. Thus inhibition of angiogenesis could prove to be a relatively tumor-selective therapeutic approach in comparison with standard cytotoxic therapies such as chemotherapy and radiotherapy. For these reasons, there has recently been great interest in identifying agents with potential antiangiogenic activity. Thalidomide was developed as a sedative in the late 1950s but was quickly removed from the market when its teratogenic effects were discovered. Recently, D’Amato et al13 demonstrated that thalidomide has potent antiangiogenic activity in vivo, a property that may be related to its mechanism of teratogenesis. In light of its excellent oral bioavailability and minimal side effects, thalidomide is a promising antiangiogenic agent for long-term therapy in patients with vascular tumors. We therefore conducted a phase II trial of high-dose thalidomide in patients with recurrent high-grade astrocytomas and mixed gliomas. PATIENTS AND METHODS
Patients and Treatment Patients were treated with thalidomide 800 mg/d, escalating by 200 mg/d every 2 weeks until a total dose of 1,200 mg/d was achieved. Dose-limiting toxicities were defined as grade 2 peripheral neuropathy, grade 3 nonhematologic toxicity, and grade 4 hematologic toxicity. Treatment cycles were defined as 8 weeks. Physical and neurologic examinations were performed every 2 weeks for the first 8 weeks and then monthly. Complete blood cell count, serum biochemistries, and beta human chorionic gonadotropin (serum pregnancy for all premenopausal women) were performed every 4 weeks for the duration of the study. Performance status was assessed every 8 weeks. Assessment of disease response was made at the end of every 8-week time interval. This included a neurologic examination and a gadolinium enhanced magnetic resonance imaging (MRI) scan.
All patients with measurable tumors were evaluated by objective criteria as proposed by Macdonald et al.14 In brief, the measurement of the enhancing mass was defined as the largest cross-sectional area, which was calculated by multiplying the largest cross-sectional diameter times the largest diameter perpendicular to it. Complete response was defined as resolution of all contrast enhancement on MRI or computed tomography scans for at least 1 month, whereas partial response was defined as a Because constipation was an expected side effect of treatment with thalidomide, aggressive prophylactic management of bowel function was routinely implemented. Antiseizure medications were allowed; however, secondary to the expected sedative effects of thalidomide, patients were not permitted to take other sedative neuroleptics such as benzodiazepines or other types of sleep medication. Secondary to the theoretical risk of increased bleeding, patients were not allowed to use nonsteroidal anti-inflammatory drugs.
Study Design
Immunoassays
Analytic Method
Pharmacokinetic Analysis RESULTS A total of 39 patients were accrued to this trial, 36 of whom were assessable for both toxicity and response. The three nonassessable patients were enrolled onto the trial but never received any thalidomide. Five additional patients received less than the initially required 3 weeks of treatment (one discontinued treatment because of drug rash, two discontinued because of seizure, and two discontinued because of tumor progression). There were a total of 20 men and 19 women accrued to this trial, with 21 patients having a KPS of 90 to 100 and 17 patients having a KPS of 70 to 80 (Table 1). Histologic subtypes included glioblastoma (n = 25), anaplastic astrocytoma (n = 12), and other anaplastic gliomas (n = 2). Thirty patients had undergone a full craniotomy with tumor resection at some point in their disease, and nine patients had biopsies only. All patients had received full-dose standard fractionated external-beam radiation therapy, and 20 patients had received prior chemotherapy (13 having received one prior treatment regimen and seven having received two prior regimens; Table 1). The interval between completing their initial fractionated radiation and beginning treatment with thalidomide ranged from 9 weeks to more than 15 years, with the median time being 50 weeks.
Thalidomide was generally well tolerated. There were four cases of grade 4 cortical toxicity, all of which were seizures (Table 2). All patients who had seizures on this trial had a prior history of seizures and/or had documented tumor progression at the time of the new seizure activity. The most common toxicities clearly related to thalidomide were easily managed constipation (one grade 1 episode and six grade 2 episodes) and somnolence (12 grade 1 episodes, three grade 2 episodes, and six grade 3 episodes). Although there were six incidences of grade 3 somnolence, most patients experienced an apparent tachyphylaxis to this side effect such that by the second or third week of treatment, the somnolence was significantly improved despite having the dose of thalidomide increased with time. One patient treated with thalidomide for almost 1 year developed a grade 2 peripheral neuropathy in his lower extremities, which only partially improved (to a grade 1 toxicity) once thalidomide was discontinued secondary to tumor progression.
Thirty-four patients who received a single oral dose of thalidomide were assessable for pharmacokinetic evaluation (Table 3). Three additional patients had pharmacokinetic data obtained, but the data were not interpretable. Using weighted nonlinear least squares as the estimator, a one-compartment model provided the best fit (Fig 1 A). The time to maximum (peak) drug concentration of thalidomide ranged between 1.67 and 8.75 hours, with a median value of 4.71 hours. The median maximum concentration of drug attained after the single oral dose was 4.09 µg/mL (range, 0.94 to 12.10 µg/mL). The oral clearance and apparent volume of distribution were 13.83 ± 7.79 L/h and 146.19 ± 92.63 L, respectively, and the elimination half-life was 8.28 ± 6.00 hours (Table 3). Thirty-one patients who received multiple daily doses of oral thalidomide were assessable for pharmacokinetic assessment (Tables 4 and 5). Using a posterior probability as the estimator, a one-compartment model provided the best fit. The oral clearance and apparent volume of distribution were 12.65 ± 6.63 L/h and 123.75 ± 72.90 L, respectively, and the elimination half-life was 8.31 ± 7.12 hours (Figs 1A, 1B, and 2). We could not demonstrate any clear relationship between thalidomide plasma levels or pharmacologic parameters and the use of concurrent medications (ie, the use of antiepileptics) (Fig 3). There did, however, seem to be an inverse correlation between age and thalidomide clearance (Fig 2).
There were four patients who had clear radiographic regression of their tumor, which amounted to partial responses according to the Levin criteria15 and two partial and two minimal responses according to the Macdonald objective criteria.14 Twelve additional patients had stable disease. Kaplan-Meier curves for TTP and overall survival for the entire group are shown in Fig 4. The median TTP was 10 weeks for the entire group, 8 weeks for nonresponders, 15 weeks for patients with stable disease, and 33 weeks for patients with radiographic responses. Median survival measured from the initiation of thalidomide treatment was 28 weeks for the study population, 74 weeks for patients with radiographic responses, 30 weeks for patients with stable disease, and 22 weeks for patients with disease progression. Although the numbers are small, there was no obvious impact of histology (anaplastic astrocytoma v glioblastoma) on likelihood of response, response duration, or survival, although patients with glioblastoma who did not respond to treatment had the shortest overall survival. In contrast, patients with glioblastoma who responded to thalidomide had similar outcomes to those of patients with anaplastic astrocytomas. Finally, the interval from initial standard fractionated radiation therapy and initiation of thalidomide treatment could not be correlated with either radiographic response, TTP, or overall survival.
Secondary to protein instability in patient blood samples, complete data for serum and urine levels of bFGF and VEGF were only available for 16 patients. The reason for this is that there were some serum and urine samples thawing while in transit between the participating institutions. Thus we chose to only use the samples from the center at which the assays were being performed (Dana-Farber Cancer Institute, Boston, MA) and only when optimal conditions for specimen handling could be documented to guarantee quality assurance. Urine and serum VEGF levels were not found to be correlated with any observed clinical parameters, including tumor response, progression, TTP, or overall survival. Despite the small number of such patients, however, there was a statistically significant relationship between changes in serum bFGF and radiographic response and survival (similar changes were also seen in urine bFGF levels, but these did not reach statistical significance because of large intrapatient variability). Specifically, patients whose serum bFGF increased by the third to seventh week of treatment (compared with baseline levels) had a shorter survival than those whose serum bFGF remained constant, whereas those patients whose serum bFGF decreased survived the longest (median survival durations of 19, 32, and 43 weeks, respectively; P < .005 by log-rank analysis). A similar relationship was seen with TTP (median TTP of 12 weeks for patients with increasing bFGF compared with 17 weeks for those with decreasing or stable bFGF; P < .02 by log-rank analysis). All patients who had radiographic and clinically stable disease had either stable or decreasing bFGF. DISCUSSION In the search for novel therapeutic targets for malignant gliomas, angiogenesis seems to be a promising candidate, given the highly vascular nature of these tumors and the growing biologic data that demonstrate overexpression of inducers and repression of natural suppressors of angiogenesis in glioma tissue.8-12,16-18 Thus the biology of astrocytoma-mediated angiogenesis seems to fit nicely into the model of tumor angiogenesis proposed by Folkman,5,6 called the "angiogenic switch," in which angiogenic homeostasis is maintained by a physiologic balance between natural mediators and inhibitors of angiogenesis.19 In contrast, tumor-associated angiogenesis constitutes an imbalance in these factors, such that the angiogenic switch is "on" rather than "off." A natural extension to the angiogenic switch model suggests that if either proangiogenic stimuli can be downregulated or inhibitors of angiogenesis can be upregulated, then the angiogenic switch might be turned off in a tumor, thereby leading to suppression of further tumor growth. In theory, angiogenesis inhibition should be a relatively nontoxic, tumor-selective therapy, given the belief that physiologic angiogenesis is limited to the female reproductive cycle and wound healing in the adult. During the past decade, there have been a number of reports in the literature demonstrating the ability of several agents with antiangiogenic activity to inhibit tumor-mediated angiogenesis and tumor growth in animals.13,18,20-26 The promise of a nontoxic, tumor-selective cancer therapy based on the rather simple idea of starving the tumor of its blood supply has led to a significant amount of interest in this approach in the lay press. The ever-increasing number of reports in the biomedical literature of new angiogenesis inhibitors is leading to an outcry for the "cure" from desperate cancer patients and their families. The reality, however, is that until recently, there have been few antiangiogenic agents suitable for human trials, and the hypothesis that angiogenesis inhibition may represent an effective therapeutic strategy against human cancer remains untested. Our trial represents one of the first attempts by the National Cancer Institute to perform a prospective, multi-institutional trial of an antiangiogenic agent in a well-defined group of patients with advanced malignant tumor. The results confirmed our expectation that thalidomide would be well tolerated, with modest sedation and mild constipation being the only consistent side effects of treatment. One might actually conclude that thalidomide was a rather toxic drug when one first looks at the adverse reaction data from this group of patients (Table 2; four episodes of grade 4 toxicity and 14 episodes of grade 3 toxicity); however, it is important to realize that this merely represents all adverse reactions, regardless of whether they were truly related to the study drug. As previously discussed, the four grade 4 toxic events were all seizures, and all patients who had seizures on this trial had a prior history of seizures and/or had documented tumor progression at the time of the new seizure activity. At least five of the grade 3 adverse events may have been related to the patients tumors (sensory and motor deficits, as well as headache). The majority of the remaining grade 3 adverse events were somnolence, and as previously discussed, most patients experienced an apparent tachyphylaxis to this side effect, such that by the second or third week of treatment, the somnolence was significantly improved despite having the dose of thalidomide increased with time. Thus most patients were not limited by treatment with thalidomide and remained fully functional throughout their treatment course if not limited by their tumors. Pharmacology data demonstrated that hepatic enzyme-inducing drugs (ie, phenytoin, phenobarbital) had little effect on the metabolism of thalidomide. Despite its acceptable toxicity and safety profile, thalidomide does not seem to have high activity in this patient population according to standard criteria for the assessment of either cytotoxic (6% radiographic objective responses) or cytostatic (10-week median TTP) activity. Nevertheless, it is clear that some patients experienced tumor regression or stabilization in association with thalidomide therapy for a prolonged period of time. This finding is intriguing given that thalidomide has no known direct antiproliferative or cytotoxic activity in vitro or in vivo (unpublished results). Thus the mechanism of the antitumor activity of thalidomide seems to be indirect, which is consistent with the hypothesis that the antitumor activity that was observed might have been mediated through an antiangiogenesis mechanism. However, it is still possible that the apparent antitumor effects of thalidomide may not have been antiangiogenic but rather were mediated through other mechanisms, such as the demonstrated ability of thalidomide to modulate the immune system.27-30 It is also conceivable that the apparent antitumor effects of thalidomide represented a radiographic artifact through which effects on the blood-brain/blood-tumor barrier lead to decreased extravasation of gadolinium contrast material into the surrounding brain tissue. Such a phenomenon, observed on computed tomography scans of patients who receive glucocorticoids, can be misconstrued as a true antitumor effect. Although it is impossible to rule out such an effect, the fact that patients with decreased enhancement had the longest survival is suggestive that these patients were experiencing a true antitumor effect rather than a radiographic artifact. There was no clear correlation between circulating levels of VEGF and any observed clinical parameter. Whether this reflects a true lack of effect of thalidomide on VEGF expression or rather that such effects are only occurring at the cellular level could not be discerned from this study. Future studies designed to acquire tumor tissue before and after thalidomide treatment will ultimately be required to answer this question. In contrast, and despite the small number of patients, the relationship between the pattern of serum bFGF changes and radiographic/clinical response, TTP, and overall survival is consistent with a biologic and potentially antiangiogenic effect from thalidomide. bFGF is a potent endothelial cell mitogen and is highly overexpressed by malignant glioma cells in vivo. Whether bFGF is directly downregulated by thalidomide or whether the decrease merely reflects an antitumor effect of treatment remains unclear. It is, however, theoretically attractive to hypothesize that downregulation of bFGF is one of the antiangiogenic mechanisms of thalidomide. Such a hypothesis is consistent with recent data that demonstrate the ability of thalidomide to inhibit bFGF-induced angiogenesis in vivo in a rabbit cornea micropocket assay.31 Future, larger prospective studies with thalidomide as well as with cytotoxic agents will be required to ultimately determine whether serum bFGF represents a useful biologic marker for glioma progression and/or antiangiogenic effect. In summary, despite the large amount of attention that the media has devoted to thalidomide, the drug seems to have only minimal antitumor activity in patients with recurrent malignant gliomas, with objective responses and/or prolonged periods of tumor stabilization noted in a relatively small proportion of patients. Thalidomide, therefore, does not seem to be useful as a single agent for most patients with recurrent high-grade gliomas. Future studies will evaluate thalidomide as a form of adjuvant therapy in newly diagnosed patients at the time of minimal residual tumor when the goal of therapy is to inhibit neovascularization of microscopic residual tumor cells to prevent the development of macroscopic, angiogenesis-dependent tumor masses. Additionally, with growing data demonstrating synergism between antiangiogenic agents and traditional cytotoxic therapies,32-34 future studies will also evaluate thalidomide in combination with chemotherapy and/or radiotherapy. Although more selective and potent antiangiogenic agents are just now entering clinical trials, antiangiogenic agents like thalidomide will likely prove most valuable as part of a multimodality anticancer strategy rather than as a single therapeutic approach. REFERENCES
1.
DeAngelis LM, Yahalom J, Thaler HT, et al: Combined modality therapy for primary CNS lymphoma. J Clin Oncol 10:635-643, 1992 2. Cairncross JG, Macdonald DR: Successful chemotherapy for recurrent malignant oligodendroglioma. Neurol 23:360-364, 1988
3.
Fine HA, Mayer RJ: Primary central nervous system lymphoma. Ann Int Med 119:1093-1104, 1993 4. Fine HA: The basis for current treatment recommendations for malignant gliomas. J Neurooncol 20:111-121, 1994[Medline] 5. Folkman J: Tumor angiogenesis: Therapeutic implications. N Engl J Med 285:1182-1186, 1971
6.
Folkman J: What is the evidence that tumors are angiogenesis dependent? Cancer Inst 82:4-6, 1990 7. Toi M, Inada K, Suzuki H, et al: Tumor angiogenesis in breast cancer: Its importance as a prognostic indicator and the association with VEGF expression. Cancer Res Treat 36:193-204, 1995 8. Plate KH, Breier G, Weich HA, et al: Vascular endothelial growth factor is a potential tumour angiogenesis factor in human gliomas in vivo. Nature 359:845-848, 1992[Medline] 9. Shweiki D, Itin A, Soffer D, et al: Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis. Nature 359:843-845, 1992[Medline] 10. Brem S, Cotran R, Folkman J: Tumor angiogenesis: A quantitative method for histologic grading. J Natl Cancer Inst 48:347-356, 1972 11. Takahashi J, Fukumoto M, Igorshi K, et al: Correlations of bFGF expression levels with degree of malignancy and vascularity in human gliomas. J Neurosurg 76:792-798, 1992[Medline]
12.
Zagzag D, Miller D, Sato Y, et al: Immunohistochemical localization of bFGF in astrocytomas. Cancer Res 50:7393-7398, 1990
13.
D’Amato RJ, Loughnan MS, Flynn E, et al: Thalidomide is an inhibitor of angiogenesis. Proc Natl Acad Sci U S A 91:4082-4085, 1994 14. Macdonald DR, Cascino TL, Schold SC Jr, et al: Response criteria for phase II studies of supratentorial malignant glioma. J Clin Oncol 8:1277-1280, 1990[Abstract] 15. Levin VA, Wilson CV, Crafts D, et al: Criteria for evaluating patients undergoing chemotherapy for malignant brain tumor. J Neurosurg 47:329-335, 1977[Medline] 16. Millauer B, Shawver LK, Plate KH, et al: Glioblastoma growth inhibited in vivo by a dominant-negative Flk-1 mutant. Nature 367:576-579, 1994[Medline] 17. Maxwell M, Naber SP, Wolfe HJ, et al: Coexpression of platelet-derived growth factor (PDGF) and PDGF-receptor genes by primary human astrocytomas may contribute to their development and maintenance. J Clin Invest 86:131-140, 1990
18.
Hsu SC, Volpert OV, Steck PA, et al: Inhibition of angiogenesis in human glioblastomas by chromosome 10 induction of thrombospondin. Cancer Res 56:5684-5691, 1996 19. Hanahan D, Folkman J: Patterns and emerging mechanisms of the angiogenesis switch during tumorigenesis. Cell 86: 353-364, 1996[Medline] 20. Fotsis T, Zhang Y, Pepper MS, et al: The endogenous oestrogen metabolite 2-methoxyoestradiol inhibits angiogenesis and suppresses tumour growth. Nature 368:237-239, 1994[Medline] 21. Ingber D, Fujita T, Kishimoto S, et al: Synthetic analogues of fumagillin that inhibit angiogenesis and suppress tumor growth. Nature 348:555-557, 1990[Medline] 22. Lee JK, Choi B, Sobel RA, et al: Inhibition of growth and angiogenesis of human neurofibrosarcoma by heparin and hydrocortisone. J Neurosurg 73:429-435, 1990[Medline] 23. O’Reilly MS, Boehm T, Shing Y, et al: Endostatin: An endogenous inhibitor of angiogenesis and tumor growth. Cell 88:277-285, 1997[Medline]
24.
Sakamoto N, Iwahana M, Tanaka NG, et al: Inhibition of angiogenesis and tumor growth by a synthetic laminin peptide, CDPGYIGSR-NH2. Cancer Res 51:903-906, 1991 25. Tanaka T, Manome Y, Wen P, et al: Viral vector-mediated transduction of a modified platelet factor 4 cDNA inhibits angiogenesis and tumor growth. Nat Med 3:1-6, 1997[Medline] 26. Tanaka T, Cao Y, Folkman J, et al: Viral-vector mediated antiangiogenic gene therapy utilizing an angiostatin cDNA. Res 58:3362-3369, 1998 27. Rovelli A, Arrigo C, Nesi F, et al: The role of thalidomide in the treatment of refractory chronic graft-versus-host disease following bone marrow transplantation in children. Transplant 21:577-581, 1998
28.
Hamuryudan V, Mat C, Saip S, et al: Thalidomide in the treatment of the mucocutaneous lesions of the Behcet syndrome: A randomized, double-blind, placebo-controlled trial. Ann Intern Med 128:443-450, 1998 29. Moller DR, Wysocka M, Greenlee BM, et al: Inhibition of IL-12 production by thalidomide. J Immunol 159:5157-5161, 1997[Abstract]
30.
Jacobson JM, Greenspan JS, Spritzer J, et al: Thalidomide for the treatment of oral aphthous ulcers in patients with human immunodeficiency virus infection: National Institute of Allergy and Infectious Diseases AIDS Clinical Trials Group. J Med 336:1487-1493, 1997 31. Kenyon BM, Browne F, D’Amato RJ: Effects of thalidomide and related metabolites in a mouse corneal model of neovascularization. Exp Eye Res 64:971-978, 1997[Medline]
32.
Teicher BA, Sotomayor EA, Huang ZD: Antiangiogenic agents potentiate cytotoxic cancer therapies against primary and metastatic disease. Cancer Res 52:6702-6704, 1992 33. Teicher BA, Holder SA, Ara G, et al: Comparison of several antiangiogenic regimens alone and with cytotoxic therapies in the Lewis lung carcinoma. Cancer Chemother Pharmacol 38:169-177, 1996[Medline] 34. Teicher BA, Holden SA, Ara G, et al: Potentiation of cytotoxic cancer therapies by TNP-470 alone and with other anti-angiogenic agents. Cancer 57:1-6, 1994 Submitted June 29, 1999; accepted September 30, 1999. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
50% reduction in the size of enhancing tumors on MRI scans. Progressive disease was defined as a 
