Improved survival outcomes with the incidental use of beta-blockers among patients with non-small-cell lung cancer treated with definitive radiation therapy.

BACKGROUND: Preclinical studies have shown that norepinephrine can directly stimulate tumor cell migration and that this effect is mediated by the beta-adrenergic receptor. PATIENTS AND METHODS: We retrospectively reviewed 722 patients with non-small-cell lung cancer (NSCLC) who received definitive radiotherapy (RT). A Cox proportional hazard model was utilized to determine the association between beta-blocker intake and locoregional progression-free survival (LRPFS), distant metastasis-free survival (DMFS), disease-free survival (DFS), and overall survival (OS). RESULTS: In univariate analysis, patients taking beta-blockers (n = 155) had improved DMFS (P < 0.01), DFS (P < 0.01), and OS (P = 0.01), but not LRPFS (P = 0.33) compared with patients not taking beta-blockers (n = 567). In multivariate analysis, beta-blocker intake was associated with a significantly better DMFS [hazard ratio (HR), 0.67; P = 0.01], DFS (HR, 0.74; P = 0.02), and OS (HR, 0.78; P = 0.02) with adjustment for age, Karnofsky performance score, stage, histology type, concurrent chemotherapy, radiation dose, gross tumor volume, hypertension, chronic obstructive pulmonary disease and the use of aspirin. There was no association of beta-blocker use with LRPFS (HR = 0.91, P = 0.63). CONCLUSION: Beta-blocker use is associated with improved DMFS, DFS, and OS in this large cohort of NSCLC patients. Future prospective trials can validate these retrospective findings and determine whether the length and timing of beta-blocker use influence survival outcomes.

Dormancy of micrometastases: balanced proliferation and apoptosis in the presence of angiogenesis suppression.

In cancer patients, dormant micrometastases are often asymptomatic and clinically undetectable, for months or years, until relapse. We have studied dormant lung metastases under angiogenesis suppression in mice. The metastases exhibited rapid growth when the inhibition of angiogenesis was removed. Tumour cell proliferation, as measured by bromodeoxyuridine incorporation and immunohistochemical staining proliferating cell nuclear antigen, was not significantly different in dormant and growing metastases. However, tumour cells of dormant metastases exhibited a more than threefold higher incidence of apoptosis. These data show that metastases remain dormant when tumour cell proliferation is balanced by an equivalent rate of cell death and suggest that angiogenesis inhibitors control metastatic growth by indirectly increasing apoptosis in tumour cells.

Angiostatin induces and sustains dormancy of human primary tumors in mice.

There is now considerable direct evidence that tumor growth is angiogenesis-dependent. The most compelling evidence is based on the discovery of angiostatin, an angiogenesis inhibitor that selectively instructs endothelium to become refractory to angiogenic stimuli. Angiostatin, which specifically inhibits endothelial proliferation, induced dormancy of metastases defined by a balance of apoptosis and proliferation. We now show that systemic administration of human angiostatin potently inhibits the growth of three human and three murine primary carcinomas in mice. An almost complete inhibition of tumor growth was observed without detectable toxicity or resistance. The human carcinomas regressed to microscopic dormant foci in which tumor cell proliferation was balanced by apoptosis in the presence of blocked angiogenesis. This regression of primary tumors without toxicity has not been previously described. This is also the first demonstration of dormancy therapy, a novel anticancer strategy in which malignant tumors are regressed by prolonged blockade of angiogenesis.

Antiangiogenic activity of the cleaved conformation of the serpin antithrombin.

Antithrombin, a member of the serpin family, functions as an inhibitor of thrombin and other enzymes. Cleavage of the carboxyl-terminal loop of antithrombin induces a conformational change in the molecule. Here it is shown that the cleaved conformation of antithrombin has potent antiangiogenic and antitumor activity in mouse models. The latent form of intact antithrombin, which is similar in conformation to the cleaved molecule, also inhibited angiogenesis and tumor growth. These data provide further evidence that the clotting and fibrinolytic pathways are directly involved in the regulation of angiogenesis.

Expression of angiostatin cDNA in a murine fibrosarcoma suppresses primary tumor growth and produces long-term dormancy of metastases.

Tumor growth and metastasis are angiogenesis dependent. Previously, we reported that angiostatin, a potent angiogenesis inhibitor, produced by a primary Lewis lung carcinoma suppressed its growth of lung metastases (O'Reilly, M.S., L. Holmgren, Y. Shing, C. Chen, R.A. Rosenthal, M. Moses, W.S. Lane, Y. Cao, E.H. Sage, and J. Folkman. 1994. Cell. 79:315-328). Now we show that a shift of balance of tumor angiogenesis by gene transfer of a cDNA coding for mouse angiostatin into murine T241 fibrosarcoma cells suppresses primary and metastatic tumor growth in vivo. Implantation of stable clones expressing mouse angiostatin in C57Bl6/J mice inhibits primary tumor growth by an average of 77%. After removal of primary tumors, the pulmonary micrometastases in approximately 70% of mice remain in a microscopic dormant and avascular state for the duration of the experiments, e.g., 2-5 mo. The tumor cells in the dormant micrometastases exhibit a high rate of apoptosis balanced by a high proliferation rate. Our study, to our knowledge, for the first time shows the diminished growth of lung metastases after removal of the primary tumor, suggesting that metastases are self-inhibitory by halting angiogenesis. Our data may also provide a novel approach for cancer therapy by antiangiogenic gene therapy with a specific angiogenesis inhibitor.

Inhibition of angiogenesis in vivo by interleukin 12.

BACKGROUND: In previous animal studies, interleukin 12 (IL 12) was shown to inhibit the growth of a wide spectrum of tumors in vivo but to have no direct effect on tumor cells in vitro. Also, contrary to the expectation of a T-cell-mediated effect, the antitumor activity of IL 12 was not completely abrogated in tests of T-cell-deficient mice. These observations suggest that IL 12 may possess antiangiogenic properties that account for its tumor-inhibitory effects in vivo. PURPOSE: Our goal was to investigate the hypothesis that IL 12 has antiangiogenic properties. METHODS: A model of basic fibroblast growth factor-induced corneal neovascularization in mice was used to evaluate the effects of IL 12 and interferon gamma (IFN gamma) on angiogenesis in vivo. Different strains of male mice, e.g., immunocompetent C57BL/6 mice, severe combined immune-deficient (SCID) mice, natural killer cell-deficient beige mice, and T-cell-deficient nude mice, were treated with IL 12 (1 microgram/day) intraperitoneally for 5 consecutive days. The extent of neovascularization in response to a basic fibroblast growth factor pellet and the inhibition of neovascularization by IL 12 or IFN gamma were assessed by measuring the maximal vessel length and the corneal circumference involved in new blood vessel formation. The antitumor activities of IL 12 and of the angiogenesis inhibitor AGM-1470 were evaluated in Lewis lung carcinoma-bearing mice. In vitro proliferation studies were performed on bovine capillary endothelial cells, mouse pancreatic islet endothelial cells, and mouse hemangioendothelioma cells. RESULTS: IL 12 treatment almost completely inhibited corneal neovascularization in C57BL/6, SCID, and beige mice. This potent suppression of angiogenesis was prevented by the administration of IFN gamma-neutralizing antibodies, suggesting that the suppression was mediated through IFN gamma. In addition, the administration of IFN gamma reproduced the antiangiogenic effects observed during treatment with IL 12. Treatment with IL 12 and AGM-1470 combined did not increase toxicity and showed a trend toward enhanced antitumor efficacy in Lewis lung carcinoma-bearing mice. CONCLUSIONS: IL 12 strongly inhibits neovascularization. This effect is not mediated by a specific cell type of the immune system. Instead, IL 12 has been shown to induce IFN gamma, which, in turn, appears to play a critical role as a mediator of the antiangiogenic effects of IL 12. IMPLICATIONS: Recognition of the mechanisms of the antiangiogenic properties of IL 12 may be crucial in planning its clinical applications, including a possibility of coadministration with other inhibitors of neovascularization.

Radiation therapy to a primary tumor accelerates metastatic growth in mice.

The surgical removal of a primary tumor can result in the rapid growth of metastases. The production of angiogenesis inhibitors by the primary tumor is one mechanism for the inhibition of metastatic tumor growth. The effect of curative radiotherapy to a primary tumor known to make an inhibitor of angiogenesis and the effects on distant metastases has not been studied. We here show that the eradication of a primary Lewis lung carcinoma (LLC-LM), which is known to generate angiostatin, is followed by the rapid growth of metastases that kill the animal within 18 days after the completion of radiation therapy. The right thighs of C57BL/6 mice (n = 25) were injected s.c. with 1 x 10(6) LLC-LM cells. Animals were randomized to one of five groups: no irradiation, 40 Gy in one fraction, 30 Gy in one fraction, 40 Gy in two 20 Gy fractions, or 50 Gy in five 10 Gy fractions. Tumors were clinically eradicated in each treatment group. All of the surviving animals became dyspneic and were killed within 14-18 days after the completion of radiation therapy. Examination of their lungs revealed >46 (range, 46-62) surface metastases in the treated animals compared with 5 (range, 2-8) in the untreated animals. The lung weights had increased from 0.2 g (range, 0.19-0.22 g) in the controls to 0.58 g (range 0.44-0.84) in the experimental animals. The most effective dose regimen was 10 Gy per fraction for five fractions, and serial experiments were conducted with this fractionation scheme. Complete response of the primary tumor was seen in 25 of 35 (71%) mice. The average weight of the lungs in the nonirradiated animals was 0.22 g (range, 0.19-0.24 g) and in the irradiated animals was 0.66 g (range, 0.61-0.70 g). The average number of surface metastases increased from five per lung (range, 2-13) in the control animals to 53 per lung (range, 46-62) in the irradiated animals. Both differences were statistically significant with P < 0.001. If the nontumor-bearing leg was irradiated or the animals were sham-irradiated, no difference in the number of surface metastases or lung weights was observed between the control group and the treated group. Urinary levels of matrix metalloproteinase 2, the enzyme responsible for angiostatin processing in this tumor model, were measured and correlated with the viability and size of the primary tumor. Administration of recombinant angiostatin prevented the growth of the metastases after the treatment of the primary tumor. In this model, the use of radiation to eradicate a primary LLC-LM tumor results in the growth of previously dormant lung metastases and suggests that combining angiogenesis inhibitors with radiation therapy may control distant metastases.

A recombinant human angiostatin protein inhibits experimental primary and metastatic cancer.

Endogenous murine angiostatin, identified as an internal fragment of plasminogen, blocks neovascularization and growth of experimental primary and metastatic tumors in vivo. A recombinant protein comprising kringles 1-4 of human plasminogen (amino acids 93-470) expressed in Pichia pastoris had physical properties (molecular size, binding to lysine, reactivity with antibody to kringles 1-3) that mimicked native angiostatin. This recombinant Angiostatin protein inhibited the proliferation of bovine capillary endothelial cells in vitro. Systemic administration of recombinant Angiostatin protein at doses of 1.5 mg/kg suppressed the growth of Lewis lung carcinoma-low metastatic phenotype metastases in C57BL/6 mice by greater than 90%; administration of the recombinant protein at doses of 100 mg/kg also suppressed the growth of primary Lewis lung carcinoma-low metastatic phenotype tumors. These findings demonstrate unambiguously that the antiangiogenic and antitumor activity of endogenous angiostatin resides within kringles 1-4 of plasminogen.

Antiangiogenic scheduling of chemotherapy improves efficacy against experimental drug-resistant cancer.

To reveal the antiangiogenic capability of cancer chemotherapy, we developed an alternative antiangiogenic schedule for administration of cyclophosphamide. We show here that this antiangiogenic schedule avoided drug resistance and eradicated Lewis lung carcinoma and L1210 leukemia, an outcome not possible with the conventional schedule. When Lewis lung carcinoma and EMT-6 breast cancer were made drug resistant before therapy, the antiangiogenic schedule suppressed tumor growth 3-fold more effectively than the conventional schedule. When another angiogenesis inhibitor, TNP-470, was added to the antiangiogenic schedule of cyclophosphamide, drug-resistant Lewis lung carcinomas were eradicated. Each dose of the antiangiogenic schedule of cyclophosphamide induced the apoptosis of endothelial cells within tumors, and endothelial cell apoptosis preceded the apoptosis of drug-resistant tumor cells. This antiangiogenic effect was more pronounced in p53-null mice in which the apoptosis of p53-null endothelial cells induced by cyclophosphamide was so vigorous that drug-resistant tumors comprising 4.5% of body weight were eradicated. Thus, by using a dosing schedule of cyclophosphamide that provided more sustained apoptosis of endothelial cells within the vascular bed of a tumor, we show that a chemotherapeutic agent can more effectively control tumor growth in mice, regardless of whether the tumor cells are drug resistant.

Generation of multiple angiogenesis inhibitors by human pancreatic cancer.

A primary inoculum of human pancreatic cancer cells (BxPC-3) has the ability to inhibit the growth of a secondary tumor in an in vivo animal model. Such ability suggests that the primary tumor is producing inhibitors that act at the site of the secondary tumor. Accordingly we attempted to discover which inhibitors are produced by pancreatic cancer cells. We determined that pancreatic cancer cells process angiostatin isoforms from plasminogen. Additionally, we isolated and characterized an uncleaved "latent" antiangiogenic antithrombin (aaAT) molecule processed from systemically available AT by pancreatic cancer cells as well as a cleaved form of aaAT processed from systemically available AT by pancreatic cancer cells. Human AT, cleaved with human neutrophil elastase, inhibits angiogenesis in the chorioallantoic membrane assay. This human aaAT molecule is able to inhibit the growth of pancreatic tumors in immune-compromised mice. Our work represents the first demonstration of multiple angiogenesis inhibitors from a single tumor and suggests that antiangiogenic therapies may provide an avenue for future treatment of pancreatic cancer.

Angiostatin: an endogenous inhibitor of angiogenesis and of tumor growth.

Angiostatin, an internal fragment of plasminogen, is a potent inhibitor of angiogenesis, which selectively inhibits endothelial cell proliferation. When given systemically, angiostatin potently inhibits tumor growth and can maintain metastatic and primary tumors in a dormant state defined by a balance of proliferation and apoptosis of the tumor cells. We identified angiostatin while studying the phenomenon of inhibition of tumor growth by tumor mass and have elucidated one mechanism for this phenomenon. In our animal model, a primary tumor almost completely suppresses the growth of its remote metastases. However, after tumor removal, the previously dormant metastases neovascularize and grow. When the primary tumor is present, metastatic growth is suppressed by a circulating angiogenesis inhibitor. Serum and urine from tumor-bearing mice, but not from controls, specifically inhibit endothelial cell proliferation. The activity copurifies with a 38 kD plasminogen fragment which we have sequenced and named angiostatin. Human angiostatin, obtained from a limited proteolytic digest of human plasminogen, has similar activities. Systemic administration of angiostatin, but not intact plasminogen, potently blocks neovascularization and growth of metastases and primary tumors. We here show that the inhibition of metastases by a primary mouse tumor is mediated, at least in part, by the angiogenesis inhibitor angiostatin.

Inhibition of tumor growth by plasminogen-related protein-B.

BACKGROUND: Various fragments of the fibrinolytic protein plasminogen can act as antiangiogenic factors and inhibit the growth of primary and metastatic tumors in mice. Plasminogen-related gene-B encodes a putative 9 kDa protein virtually identical to the plasminogen N-terminal activation peptide, a 77-amino acid motif that is liberated from the parent plasminogen molecule during conversion to the serine proteinase plasmin. Previous data have documented enhanced transcription of plasminogen-related gene-B in neoplastic tissues. MATERIALS AND METHODS: We have tested the effects of recombinant versions of plasminogen-related protein-B and the plasminogen N-terminal activation peptide on the growth of tumors in mice, employing murine tumor cell lines implanted subcutaneously. RESULTS: The recombinant plasminogen-related protein-B significantly inhibited the growth of primary tumors in mice, while recombinant plasminogen N-terminal activation peptide elicited only a slight inhibition of tumor growth. CONCLUSION: These data suggest that plasminogen-related protein-B may have utility as a novel cancer therapeutic.

Treatment of murine hemangioendotheliomas with the angiogenesis inhibitor AGM-1470.

Hemangioma and other angiomatous diseases of childhood are common. Although most lesions are harmless and self-limiting, some are associated with significant morbidity and may be life-threatening. Interferon-alpha, a weak angiogenesis inhibitor, recently has been found to significantly reduce the mortality rate associated with life-threatening hemangiomas. The effectiveness of AGM-1470, a potent inhibitor of angiogenesis derived from the fungal product fumagillin, was tested in a mouse model of hemangioendothelioma. Thirty syngeneic mice were implanted with cells derived from a spontaneous mouse hemangioendothelioma. Tumors formed within 2 to 3 days, and the animals were then treated systemically with AGM-1470 or with saline and vehicle alone. After 22 days, the tumor volume in the saline-treated mice was 7368 +/- 2723 mm3, versus 709 +/- 73 mm3 in the mice that received AGM-1470 (P < .001). Survival was prolonged for the AGM-1470-treated mice, and there was no evidence of drug-related toxicity. All experiments were repeated. In this study, AGM-1470 was safe and highly effective in the treatment of hemangioendothelioma. AGM-1470, and other antiangiogenic agents, may provide safe and effective treatment for hemangioma and other angiomatous diseases.

The preclinical evaluation of angiogenesis inhibitors.

Angiogenesis is a fundamental process which is required for a number of physiological and pathophysiological processes. The field of angiogenesis therefore has many therapeutic implications and has progressed rapidly. Many strategies have been devised to regulate angiogenesis and several endogenous and synthetic inhibitors of angiogenesis have now been identified. These inhibitors can be used to treat a number of angiogenesis-dependent diseases and they offer a novel means of potently inhibiting tumor growth without significant toxicity or drug resistance. Recently, some of these inhibitors have entered clinical trials. In this article, I will review methods currently employed in the preclinical evaluation of angiogenesis inhibitors and I will discuss some of the implications of angiogenesis research.

Suppression of tumor growth with recombinant murine angiostatin.

Angiostatin, a 38 kDa internal fragment of plasminogen, is an antiangiogenic endothelial cell inhibitor. It regresses several primary and metastatic tumors in mice. To produce recombinant angiostatin for further structural and functional studies, the mouse angiostatin gene preceded by a sequence including a signal peptide of plasminogen was introduced into baculovirus. Recombinant murine angiostatin was purified from the culture medium of angiostatin baculovirus-infected insect cells (yield = 1 mg/liter) with a single-step of lysine-Sepharose chromatography. The angiostatin baculovirus-infected insect cells expressed and secreted a 52 kDa polypeptide that demonstrated all of the biological activities of angiostatin. A partial amino acid sequence of the NH2-terminus of the secreted protein revealed that the signal peptide was recognized and properly cleaved in insect cells. The recombinant murine angiostatin potently inhibited the proliferation of bovine capillary endothelial cells in vitro (half maximal inhibition = 50 ng/ml) and suppressed the growth of primary Lewis lung carcinoma in vivo (6 mg/kg/day, T/C = 0.08).

Zinc-binding of endostatin is essential for its antiangiogenic activity.

Endostatin is a potent angiogenesis inhibitor in vitro and in vivo. We used the yeast Pichia pastoris to express and purify soluble endostatin. It was discovered that metal chelating agents can induce N-terminal degradation of endostatin. We theorized that a metal was removed from endostatin which changed the conformation and allowed a contaminating protease to degrade the N-terminus. Atomic absorption and amino acid analysis of endostatin purified from Pichia pastoris and mammalian cells showed a 1:1 molar ratio of Zn2+ to protein. Ding et al. have shown that histidines 1, 3, 11, and aspartic acid 76 coordinate the Zn2+ atom (1). An H1/3A double, an H11A, and a D76A single mutant of endostatin were not able to regress Lewis lung carcinoma. We conclude that the ability of endostatin to bind Zn2+ is essential for its antiangiogenic activity.

Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance.

Acquired drug resistance is a major problem in the treatment of cancer. Of the more than 500,000 annual deaths from cancer in the United States, many follow the development of resistance to chemotherapy. The emergence of resistance depends in part on the genetic instability, heterogeneity and high mutational rate of tumour cells. In contrast, endothelial cells are genetically stable, homogeneous and have a low mutational rate. Therefore, antiangiogenic therapy directed against a tumour's endothelial cells should, in principle, induce little or no drug resistance. Endostatin, a potent angiogenesis inhibitor, was administered to mice bearing Lewis lung carcinoma, T241 fibrosarcoma or B16F10 melanoma. Treatment was stopped when tumours had regressed. Tumours were then allowed to re-grow and endostatin therapy was resumed. After 6, 4 or 2 treatment cycles, respectively, no tumours recurred after discontinuation of therapy. These experiments show that drug resistance does not develop in three tumour types treated with a potent angiogenesis inhibitor. An unexpected finding is that repeated cycles of antiangiogenic therapy are followed by prolonged tumour dormancy without further therapy.

Regulation of angiostatin production by matrix metalloproteinase-2 in a model of concomitant resistance.

We have previously reported the identification of the endogenous angiogenesis inhibitor angiostatin, a specific inhibitor of endothelial cell proliferation in vitro and angiogenesis in vivo. In our original studies, we demonstrated that a Lewis lung carcinoma (LLC-LM) primary tumor could suppress the growth of its metastases by generating angiostatin. Angiostatin, a 38-kDa internal fragment of plasminogen, was purified from the serum and urine of mice bearing LLC-LM, and its discovery provides the first proven mechanism for concomitant resistance (O'Reilly, M. S., Holmgren, L., Shing, Y., Chen, C., Rosenthal, R. A., Moses, M. A., Lane, W. S., Cao, Y., Sage, E. H., and Folkman, J. (1994) Cell 79, 315-328). Subsequently, we have shown that systemic administration of angiostatin can regress a wide variety of malignant tumors in vivo. However, at the time of our initial discovery of angiostatin, the source of the protein was unclear. We hypothesized that the tumor or stromal cells might produce an enzyme that could cleave plasminogen sequestered by the primary tumor into angiostatin. Alternatively, we speculated that the tumor cells might express angiostatin. By Northern analysis, however, we have found no evidence that the tumor cells express angiostatin or other fragments of plasminogen (data not shown). We now report that gelatinase A (matrix metalloproteinase-2), produced directly by the LLC-LM cells, is responsible for the production of angiostatin, which suppresses the growth of metastases in our original model.

Kringle domains of human angiostatin. Characterization of the anti-proliferative activity on endothelial cells.

Recently we have identified angiostatin, an endogenous angiogenesis inhibitor of 38 kDa which specifically blocks the growth of endothelial cells (O'Reilly, M. S., Holmgren, L., Shing, Y., Chen, C. , Rosenthal, R. A., Moses, M., Lane, W. S., Cao, Y., Sage, E. H., and Folkman, J. (1994) Cell 79, 315-328; Folkman, J. (1995) Nat. Med. 1, 27-31). Angiostatin was shown to represent an internal fragment of plasminogen containing the first four kringle structures. We now report on the inhibitory effects of individual or combined kringle structures of angiostatin on capillary endothelial cell proliferation. Recombinant kringle 1 and kringle 3 exhibit potent inhibitory activity with half-maximal concentrations (ED50) of 320 nM and 460 nM, respectively. Also, recombinant kringle 2 displays a significant inhibition, although decreased compared with both kringle 1 and kringle 3. In contrast, kringle 4 is an ineffective inhibitor of basic fibroblast growth factor-stimulated endothelial cell proliferation. Among the tandem kringle arrays, the recombinant kringle 2-3 fragment exerts inhibitory activity similar to kringle 2 alone. However, relative to kringle 2-3, a marked enhancement in inhibition is observed when individual kringle 2 and kringle 3 are added together to endothelial cells. This implies that it is necessary to open the cystine bridge between kringle 2 and kringle 3 to obtain the maximal inhibitory effect of kringle 2-3. An increased (<2-fold) inhibitory activity is observed for the kringle 1-3 fragment (ED50 = 70 nM) compared with kringle 1-4 (ED50 = 135 nM). These data indicate that the anti-proliferative activity of angiostatin on endothelial cells is shared by kringle 1, kringle 2, and kringle 3, but probably not by kringle 4 and that more potent inhibition results when kringle 4 is removed from angiostatin. Thus, in view of the variable lysine affinity of the homologous domains, it would appear that lysine binding capability does not correlate with the relative inhibitory effects of the kringle-containing constructs. However, as we also demonstrate, appropriate folding of kringle structures is essential for angiostatin to maintain its full anti-endothelial activity.

Endostatin: an endogenous inhibitor of angiogenesis and tumor growth.

We previously identified the angiogenesis inhibitor angiostatin. Using a similar strategy, we have identified endostatin, an angiogenesis inhibitor produced by hemangioendothelioma. Endostatin is a 20 kDa C-terminal fragment of collagen XVIII. Endostatin specifically inhibits endothelial proliferation and potently inhibits angiogenesis and tumor growth. By a novel method of sustained release, E. coli-derived endostatin was administered as a nonrefolded suspension. Primary tumors were regressed to dormant microscopic lesions. Immunohistochemistry revealed blocked angiogenesis accompanied by high proliferation balanced by apoptosis in tumor cells. There was no toxicity. Together with angiostatin data, these findings validate a strategy for identifying endogenous angiogenesis inhibitors, suggest a theme of fragments of proteins as angiogenesis inhibitors, and demonstrate dormancy therapy.