Apoptotic Tumor Cell Death Induced by Estramustine in Patients with Malignant Glioma1

Vol. 4, 87-91, January 1998 Clinical Cancer Research 87

Christina Valibo, A. Tommy Bergenheim, effects in human glioma cells in vitro (1, 2) and in a rat glioma Per Bergstrom, Per-Olof Gunnarsson, and model (3). The effects seem to involve the microtubuli system, resulting in an arrest of glioma cells in G2-M phase of the cell Roger Henriksson2 cycle (1, 4, 5). In addition, other cytotoxic effects involving cell Departments of Oncology [C. V., P. B., R. H.] and Neurosurgery [A. T. B.], UmeA University, 5-901 85 Ume#{225},Sweden, and Pharmacia membrane-coupled events (6-8) and DNA (2, 8) are involved. Upjohn, Lund, Sweden [P-O. G.] The phenomena of apoptosis, programmed cell death, has gained increased interest in experimental tumor research, and recently, it was suggested that EaM causes programmed cell ABSTRACT death in an experimental glioma model (9). In the present study, Estramustine phosphate (EMP), a cytotoxic drug used we have been able to demonstrate that EaM induces apoptosis in in the treatment of prostatic carcinoma, is metabolized and malignant glioma in the clinical situation. exerts specific cytotoxic effects in malignant glioma in vitro and in vivo. In the present study, we have evaluated the cytotoxic effect of EMP in the clinical situation with regard MATERIALS AND METHODS to appearance of DNA damage and its correlation to the Patients. Eighteen patients with supratentorial astrocy- uptake of estramustine (EaM) in human malignant astrocytoma grade III-IV operated on at the Department of Neurosur- toma tissue. Ten patients were given 280 mg of EMP p.o. gery were included in the study. The mean age was 54 years 12 h before surgery. Specimens from brain tumor tissue (range, 21-77 years), and the male:female ratio was 1 1 :7. Ten were collected during surgery and used for detection of patients were given 280 mg of EMP p.o. The time was chosen fragmented DNA, a hallmark of apoptosis, with in situ end from earlier pharmacokinetic studies (10), and the dose of EMP labeling (ISEL) and agarose gel techniques. The main me- is routinely used in the treatment of patients with advanced tabolite of EMP in glioma tissue, EaM, was analyzed with prostatic carcinoma. Exclusion criteria were cardiac failure, gas chromatography. It was demonstrated that EMP in- angina pectoris, and impaired liver or renal function. Eight duced clusters of ISEL-positive tumor cells and fragmenta- patients served as controls and did not receive EMP. All patients tion of DNA on agarose gels in patients treated with EMP. In routinely received 8 mg of betamethason i.v. 12 and 2 h before the same patients, a significant uptake of EaM in tumor surgery. The study was approved by the local ethical committee, tissue was demonstrated. In control patients, who were not and informed consent was obtained individually from each treated with EMP, and in two EMP-treated patients with no patient. uptake of EaM, no signs of fragmented DNA and only a few At surgery, specimens of tumor tissue were collected for scattered ISEL-positive cells were seen in the tumor tissue. routine histopathological analysis and for the analysis described Signs of apoptosis were also seen in two different experi- below. The interval between administration of EMP and sample mental models, i.e., in vitro cell cultures of rat glioma cells taking was approximately 12 h. Specimens for analysis were and an in vivo rat glioma model. It is suggested that EaM can removed with special attention to avoid denaturation. The sam- induce apoptosis by a direct effect on a subpopulation of ples were immediately fixated in formalin or frozen and stored glioma cells in human brain tumors in the clinical situation. at -70#{176}Cuntil analyzed. In Vitro Experiments. A nitrosourea-induced rat glioma INTRODUCTION cell line BT4C (kindly provided by Prof. Rolf Bjerkvig, Uni- EaM,3 a fusion molecule between l73-estradiol and nor- versity of Bergen, Bergen, Norway) was used for this study to nitrogen mustard, has been shown to exert specific cytotoxic evaluate direct effects on single cells. The tumor cells were cultured in DMEM (Fbowlab, Glasgow, United Kingdom) supplemented with 10% FCS and 60 mg/biter gentamicin at 37#{176}C with 5% CO2 for at least 3 days, until the cell growth was exponential. The medium was changed every second or third Received 5/8/97; revised 9/1 1/97; accepted 9/23/97. day. EMP (Estracyt; Kabi-Pharmacia, Berger, Sweden) was The costs of publication of this article were defrayed in part by the dissolved in sterile water and diluted in DMEM to give the payment of page charges. This article must therefore be hereby marked concentration 20 p.g/mI. The cells were exposed to EMP for 1, advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 2, 3, 4, 8, 24, and 96 h before DNA integrity analysis. In

I Supported by grants from the Swedish Cancer Society, Lions Cancer addition, cells treated as described above were spun down and Research Foundation (UmeA, Sweden), and Lundberg’s Research Foun- analyzed using the ISEL technique. dation (Gothenburg, Sweden). Rat Glioma Model. A rat glioma model using inbred 2 To whom requests for reprints should be addressed. Phone: 46-90- BD-IX rats and the BT4C glioma cell line was set up. The cells 7850000; Fax: 46-90-7852031.

3 The abbreviations used are: EaM, estramustine; EMP, estramustine were grown as a monolayer for 1 week before implantation. phosphate; ISEL, in situ end labeling. Cells growing in log phase were harvested and trypsinized

Table 2 ISEL and expression of p53, bcl-2, and c-myc in eight ,‘ control patients with astrocytoma grade III-IV

ISEL p53 bcl-2 c-myc Patient no. (%) (%) (%) (%)

I 0 10 0 0 2 0 0 0 0 3 1 8 0 0 4 0 5 0 0 5 0 60 0 0 6 3 2 0 0 7 0 0 0 0 8 0 0 0 0

Quantitative analysis of ISEL

Fig. I Clusters of apoptotic cells in malignant glioma analyzed using the ISEL technique in patients treated with EMP (X80). 600

0 Table 1 Uptake of EaM. the main metabolite of EMP. in 10 patients ‘C 500 with astrocytoma grade III-IV and the percentage of tumor cells U, U) positive for ISEL and immunohistochemistry (p53. bcl-2. and c-myc) a 400 Tumor Serum U Patient EaM EaM T:S ISEL p53 bcl-2 c-myc no. (ng/g’) (ngIg) EaM” (%) (%) (%) (%) p300

I 0 0 * 0 0 0 0 -J Lu 200 2 13 0 * 40 5 0 0 Cl) 3 138 10 14 50 0 0 4 0 4 26 0 5 35 0 0 5 100 5 125 6 21 5 0 0 0 6 50 2 25 30 0 0 0 E 7 15 2 7.5 10 0 0 0 C

8 0 0 * 0 5 0 0 10 - C’,) C”) 9 86 4 22 22 0 0 0 0 10 70 5 14 24 0 0 0 hours

Fig. 2 Quantitative analysis of ISEL-positive apoptotic cells in a rat Mean 52.3 2.9 17.5 glioma model in vivo at various time points after EaM treatment.

“ 1:5, tumor:serum ratio; , not applicable.

totic at this time (12, 13). The number of ISEL-positive cells before being spun down and diluted in MEM supplemented with was quantified in the light microscope. Five randomly chosen

5% BD-IX rat serum to give 20,000 cells/S p.1 ‘. The rats, 8-14 areas (5 X l0- cm2) in each section were counted. Sections weeks old, were anesthetized by i.p. administration of 1.8 mu were also processed for routine staining with H&E.

kg of a 1 : 1 : 1 mixture of fluanisonum ( 10 mg/mb ‘ ), fentan- Analysis of DNA Integrity. Paraffin sections of human

ylum [0.2 mg/m1 ‘ (Hypnorm)], and midazolam [5 mg/ml glioma specimens were deparaffinated and rehydrated according (Dormicum)]. The cells were transplanted under stereotactic to standard histological procedures. The tissue fragments were conditions in the caudate nucleus. EaM was dissolved in ethanol then suspended in digestion buffer and treated with DNase-free and castor oil to a concentration of 10 mg/mb. Twenty-four days RNase. The samples were loaded into agarose gel containing after implantation, tumor samples were taken out 0.5, 1 , 2, 4, 8, ethidium bromide. Electrophoresis was performed, and the gels 24, and 96 h after administration of EaM (20 mg/mb, i.p.). were viewed by transillumination with UV light and photo- Samples of tumor and normal brain were treated as described for graphed. X-HindIII standard was used as molecular size stand- each method. The study was approved by the local ethical ard. The analysis was performed according to the method de- committee. scribed previously (9, 14). ISEL. Samples of tumor tissue from each patient and rat Immunohistochemistry (p53, bcl-2, and c-myc). The glioma tissue were fixed in formalin and embedded in paraffin. deparaffinated sections were heated in a microwave in 0.01 M In situ detection of apoptotic cells using nucleotides of which citrate buffer (pH 6), and incubation with the primary antibodies one is labeled [AlP, dGTP, CICTP, and biotin dUTP (Boehringer [p53 (Ab6; clone do-b ; Oncogene Science, Uniondale, NY), Mannheim, Mannheim, Germany)] and DNA-polymerase 1 bcl-2 (clone 124; Cambridge Biotechnology, Cambridge, MA), (Sigma, St. Louis, MO) was performed according to the protocol and c-myc (clone 9E10; Cambridge Biotechnology)] was con- by Wijsman et a!. ( I 1 ). For details, see Vailbo et a!. (9). Normal ducted overnight. After incubation with a biotinylated secondary rat prostate 3 days after castration was used as a positive control, antibody, the sections were incubated with avidin-biotin com- because about 4% of the epithebial cells are known to be apop- plex reagents and peroxidase substrate for development. The

I 2 3 4 5 6 7 8 9 1011 121314 15161718

.-.

Fig. 3 DNA integrity analysis for detection of apoptosis in normal brain and glioma tissue in rats treated with EaM and controls. Lanes 1 and 2, 0.5 h after EaM; Lanes 3 and 4, 1 h after EaM; Lanes 5 and 6, 2 h after EaM; Lanes 7 and 8, 4 h after EaM; Lanes 9 and 10, 8 h after EaM; Lanes 11 and 12, 24 h after EaM; Lanes 13 and 14, 96 h after EaM; Lanes 15 and 16, untreated : controls; Lanes 17 and 18, rats . ‘:$ treated with drug vehicle only. ..-, . .;-. . For all lanes: odd numbers, nor- mab tissue; even numbers, tumor :;: : tissue.

sections were counterstained with Mayer’s hematoxylin sobu- I 23456789 tion. Quantification of immunostained tumor cells was made in light microscopy and expressed as the percentage of the total number of cells in five randomly chosen areas. Positive controls w-9 were from human lymphoid tissue and normal prostatic cells. Analysis of EaM. EaM, the main metabolite of EMP, was analyzed in tumor specimens and serum, using gas chromatography according to the method of Andersson et a!. (15, 16). Tissue samples were homogenized in methanol, the extracts were evaporated, and the residues were dissolved in water, extracted with hexane, and then separated on an aluminum column. The samples were dissolved in xybene and quantified by gas chromatography with NP detection (HP 5890; Hewlett Packard). Fig. 4 DNA integrity analysis for detection of apoptosis in cultured rat RESULTS glioma cells treated with EaM. Lane 1. 1-h treatment with EaM; Lane 2, 2-h treatment with EaM; Lane 3. 3-h treatment with EaM; Lane 4, 4-h Apoptosis was detected in 8 of 10 patients who were given treatment with EaM; Lane 5, 8-h treatment with EaM; Lane 6, 24-h EMP before surgery. Analyzed using the ISEL technique, clus- treatment with EaM; Lane 7, 48-h treatment with EaM; Lane 8, 96-h ters of apoptotic cells were seen in the tumors (Fig. 1). The size treatment with EaM; Lane 9, control. of the clusters varied from 100 to 500 positive cells per cluster, and the percentage of stained cells varied from S to 50% in tumor tissue (Table 1). In 2 of 10 patients treated with EMP, we detected in the tumor tissue or in the serum. Notably, these two could not detect any ISEL-positive cells. However, in these two patients were also negative for ISEL, with no signs of frag- patients, no EaM was detectable in the tumor tissue. In the mented DNA. control group, which had not been treated with EaM, only a few In the rat glioma model, EaM induced time-dependent scattered ISEL-positive cells were seen in tumor tissue (Table apoptosis in tumor tissue, but the normal brain was without 2). Using gel separation, a ladder of fragmented DNA was signs of an apoptotic cell death (Figs. 2 and 3). detected in tumor tissue from all patients treated with EMP who In vitro, EaM caused clearly visible signs of DNA frag- also showed clusters of ISEL-positive cells in their tumors. The mentation, analyzed by both DNA integrity and ISEL methods, results of the immunostaining of p53, bcl-2, and c-myc are at 4 and 8 h but not at 1, 2, 3, 24, 48, or 96 h after treatment presented in Tables 1 and 2. The analysis of EaM in tumor tissue (Fig. 4). is summarized in Table 1 . In eight patients, high levels of EaM were detected in serum as well as in the tumor tissue. The most DISCUSSION striking feature was the high EaM level in tumor tissue relative The present study demonstrates that EaM is abbe to induce to serum. In two patients (patients 1 and 8), no drug could be apoptosis in a subpopulation of cells in malignant human gli-

oma. This finding is strengthened by the observation that in the verse pattern of cytotoxic effects in glioma cells that, in the light same patients in which apoptosis was induced, a significant of current knowledge, may be features of the controlled process uptake of EaM was demonstrated. To our knowledge, this is the of programmed cell death. Effects on membrane-coupled events first report on induction of apoptosis in malignant brain tumors such as a rapid influence on ion fluxes (6, 7, 20) and formation in a clinical setting. The probability of EaM-induced apoptosis of bleb-like projections have been observed (6). Bleb formation was also supported by an earlier study in a rat glioma model that has been correlated with loss of cell viability and may be related demonstrated a time-dependent response without any signs of to lipid peroxidation via free oxygen radicals (26, 27). The inflammatory reaction or necrosis (9). In glioma culture, the observation of a rapid induction of free oxygen radicals after same time-dependent induction of apoptosis was found. EaM treatment (8) must also be encountered, because free Fragmentation of genomic DNA into low molecular weight radicals are proposed to be involved in the late steps in the DNA (< 1000 bp) is considered a sign of apoptosis (13, 17). pathway beading to apoptosis (28). ISEL is thought to facilitate a proper detection of this frag- It has been reported that p53 is frequently overexpressed in mented DNA in tissue (1 1), using the fact that double-stranded human astrocytoma of all grades (29, 30), and in half of these DNA fragments produce 3’ hydroxyl ends of terminal de- cases, the wild-type p53 was present (3 1). If a mutant form of oxynucleotides during programmed cell death. These ends can p53 is present, it will indirectly stimulate an increase of uncon- be labeled with biotinylated deoxynucleotides and can thus be trolled growth. Using immunohistochemistry, p53 is detected identified ( 1 1 , 18. 19). The pattern of DNA degradation with mainly in its mutant form (30), which must be considered when stereotypic nucleosomal-sized fragment together with the dem- evaluating the present results. In patients not treated with EMP, onstration of ISEL-positive cells strengthens the finding that a we found a clear expression of p53 (cut-off bevel, > 10%) in significant fraction of the cells in the treated human tumors die only two patients and in none of the patients treated with EMP. of apoptosis. Our finding is further supported by a previous The number of patients was limited, however, and the findings report on fragmentation of DNA after exposure to EaM (20). do not exclude the presence of wild-type p53 and subsequently Other effects on DNA. such as hampering of synthesis after are not contradictory to the demonstrated induction of apoptosis. EaM exposure, have also been demonstrated (2). In conclusion, the antimitotic drug EaM was detected at Because EaM has the capacity to induce signs of apoptosis high levels in malignant glioma tissue in a majority of patients. in vitro, it is plausible to propose direct effects on individual In tumors with a demonstrated uptake of EaM, visible damage cells to be involved. In the present study, EaM penetrates the of DNA indicative of apoptosis was encountered. This induction blood-tumor barrier and also accumulates in tumor tissue in the of apoptosis might primarily be a membrane- or cytoskeletal- majority of patients. Therefore, a direct influence of glioma triggered programmed cell death rather than a direct interaction cells, even in the clinical situation, may be assumed. It is with the DNA and thus does not necessarily contradict the noteworthy that in two patients in our study without any uptake earlier proposed mechanism of action of EaM, which involves of EaM in tumor tissue, no signs of apoptosis were detected. The the microtubular system. This finding of induced apoptosis in reason for this back of uptake is unclear, but it may be due to a the clinical setting further increases the interest in EaM and concomitant intake of calcium-containing food and/or antacid similar drugs in the treatment of malignant glioma. (21. 22). Another significant finding was that normal brain tissue had no signs of apoptosis after EaM treatment in a rat glioma REFERENCES model. In this context, it is of interest to recall that EaM affects I. von Schoultz, E., Lundblad, D., Bergh, J., Grankvist, K., and microtubule integrity and displayed toxicity only in glioma cells Henriksson, R. Estramustine binding protein and anti-proliferative effect and not in normal astrocytes (23). This selective sensitivity in of estramustine in human glioma cell lines. Br. J. Cancer, 58: 326-329, glioma cells may be explained by the high expression of a 1988. specific binding protein, EaM-binding protein, in glial tumors 2. Piepmeier. J. M., Keefe, D. L., Weinstein. M. A., Yoshida, D., Zielinski, J., Lin, 1. T., Chen, Z., and Naftolin, F. Estramustine and compared to normal brain tissue (1, 24. 25). estrone analogs rapidly and reversibly inhibit deoxyribonucleic acid Obviously, EaM caused apoptosis only in a subpopulation synthesis and alter morphology in cultured human glioblastoma cells. of glioma cells in both the present clinical situation and in rats. Neurosurgery (Baltimore), 32: 422-431, 1993. An implication from this, observed clusters of apoptotic cells, 3. Bergenheim, A. T., Elfverson, J., Gunnarsson, P-O., Edman, K., can be attributed to the well-known heterogeneity and chemore- Hartman. M., and Hennksson, R. Cytotoxic effect and uptake of estramustine in a rat glioma model. Int. J. Oncol., 5: 293-299, 1994. sistance of malignant glioma. Hence, a subpopulation of glioma 4. Dahll#{246}f,B., Billstrdm, A., Cabral, F., and Hartley-Asp, B. Estramus- cells sensitive to chemotherapeutics such as EaM may in some tine depolymerizes microtubules by binding to tubulin. Cancer Res., 53: sense be primed to undergo apoptosis. They just need to be 4573-4581, 1993. exposed to the agent. Consistent with this, a significant number 5. Bergenheim. A. 1., Zackrisson, B., Ebfverson. J., Roos. G., and of tumor cells may be resistant to chemotherapy, that is, the lack Henriksson, R. Radiosensitizing effect of estramustine in malignant of or altered signal pathway beading to apoptosis, and thus glioma in vitro and in vivo. J. Neuro-oncol., 23: 191-200, 1995. contribute to the poor prognosis in patients with malignant 6. Engstrom. K. G., Grankvist, K., and Henriksson, R. Early morpho- glioma. The cluster appearance of apoptotic cells might also be logical detection of estramustine cytotoxicity measured as alteration in cell size and shape by a new technique of microperifusion. Eur. J. due to an uneven blood supply and penetration of EaM in the Cancer, 10: 1288-1295, 1991. tumor tissue. This tumor heterogeneity regarding the induction 7. Sandstr#{246}m, P-E., Jonsson, 0., Grankvist, K., and Henriksson, R. of apoptosis needs to be further investigated. Identification of potassium flux pathways and their role in the cytotox- Earlier studies with EaM have shown a complex and di- icity of estramustine in human malignant glioma. prostatic carcinoma,

and pulmonary carcinoma cell lines. Eur. J. Cancer, 30A: 1822-1826. oxynucleotidyl transferase and nick translation assays. Cancer Res., 53: 1994. 1945-1951, 1993. 8. Henriksson, R., Bjermer, L., von Schoultz, E., and Grankvist, K. The 20. von Schoultz, E., Grankvist, K., Gustafsson, H., and Henriksson, R. effect of estramustine on microtubule is different from the direct action Effects ofestramustine on DNA and cell membrane in malignant glioma via oxygen radicals on DNA and cell membrane. Anticancer Res., 10: cells. Acta Oncob., 30: 719-723, 1991. 303-310, 1990. 21. Gunnarsson, P. 0., Davidsson, T., Andersson, S. B., Backman, C.. 9. Vallbo, C., Bergenheim, T., Bergh, A., Grankvist, K., and and Johansson, S. A. Impairment of estramustine phosphate absorption Henriksson, R. DNA fragmentation induced by the antimitotic drug by concurrent milk and food intake. Eur. J. Clin. Pharmacol., 38: estramustine in malignant rat glioma but not in normal brain, suggesting 189-193, 1990. an apoptotic cell death. Br. J. Cancer, 71: 717-720, 1995. 22. Bergenheim, A. 1., Gunnarsson, P. 0., Edman, K., von-Schoultz, 10. Gunnarsson, P. 0., Andersson, S-B., Johansson, S-A., Nilsson, C., E.. Hariz, M. I., and Henriksson, R. Uptake and retention of estramus- and Plym-Forsell, G. Pharmacokinetics of estramustine phosphate in tine and the presence of estramustine binding protein in malignant brain prostatic cancer patients. Eur. J. Clin. Pharmacol., 26: 113-119, 1984. tumors in humans. Br. J. Cancer. 67: 358-361, 1993. 11. Wijsman, J. H., Jonker, R. R., Keijzer, R., Develde, D. J. H., 23. Yoshida, D., Cornell-Bell, A., and Piepmeier, J. Selective antimi- Comelisse, C. J., and van Dierendonck, J. H. A new method to detect totic effects of estramustine correlate with its antimicrotubule properties apoptosis in paraffin sections: in situ end-labeling of fragmented on gliobbastoma and astrocytes. Neurosurgery (Baltimore). 34: 863- DNA. J. Histochem. Cytochem., 41: 7-12, 1993. 868, 1994. 12. English, H. F., Kyprianou, N., and Isaacs, J. T. Relationship be- 24. von Schoultz, E., Bergenheim, A. T., Grankvist, K., and Henriksson, R. tween DNA fragmentation and apoptosis in the programmed cell death Estramustine-binding protein in human brain tumor tissue. J. Neuro- in the rat prostate following castration. Prostate, 15: 233-250, 1989. surg., 74: 962-964, 1991. 13. Dive, C., and Wyllie, A. H. Apoptosis and cancer chemotherapy. In: 25. Bergenheim. A. T., Bj#{246}rk,P.,Bergh, J.. von Schoultz, E., Svedberg. J. A. Hickman and T. T. Tritton (eds.), Frontiers in Pharmacology: H., and Henriksson, R. Estramustine-binding protein and specific bind- Cancer Chemotherapy. pp. 21-56. Oxford, United Kingdom: Blackwell ing of the antimitotic compound estramustine in astrocytoma. Cancer Scientific, 1993. Res., 54: 4974-4979, 1994. 14. Tounekti, 0., Pron, G., Belehradek, J., Jr., and Mir, L. M. Bleomy- 26. Noseda, A., White, J. G., Godwin, P. L., Jerome, W. G., and cm, an apoptosis-mimetic drug that induces two types of cell death, Modest, E. J. Membrane damage in leukemic cells induced by ether and depending on the number of molecules internalized. Cancer Res., 53: ester lipids: an electron microscopic study. Exp. Mob. Pathol., I: 69-83, 5462-5469, 1993. 1989. 15. Andersson, S-B., Gunnarsson P. 0., Nilsson T., and Plym-Forsell, 27. Noronha-Dutra, A. A., Steen-Dutra, E. M., and Woolf, N. Epineph- G. Metabolism of estramustine phosphate in patients with prostatic rime-induced cytotoxicity of rat plasma: its effects on isolated cardiac carcinoma. Eur. J. Drug Metab. Pharmacokinet., 6: 149-154, 1981. myocytes. Lab. Invest., 6: 817-823, 1988. 16. Andersson, S-B., Lundgren, R., and Svensson, L. Gas chromato- 28. Hockenbery, D. M., Oltvai, Z. N., Yin, X-M., Milliman, C. L., and graphic determination of four metabolites of estramustine phosphate in Korsmeyer, S. J. Bcl-2 functions in an antioxidant pathway to prevent plasma. Acta Pharm. Suec., 19: 1-10, 1982. apoptosis. Cell, 75: 241-251, 1993. 17. Isaacs, J. T. Advances and controversies in the study of pro- 29. Louis, D. N. The p513 gene and protein in human brain tumors. grammed cell death/apoptosis in the development of and therapy for J. Neuropathol. Exp. Neurol., 53: 1 1-2 1,1994. cancer. Curr. Opin. Oncol., 6: 82-89, 1994. 30. Danks, R. A., Chopra. G., Gonzales, M. F., Orian, J. M., and Kaye. 18. Gavrieli, Y., Sherman, Y., and Ben-Sasson, S. A. Identification of A. H. Aberrant p53 expression does not correlate with the prognosis in programmed cell death in situ via specific labeling of nuclear DNA anaplastic astrocytoma. Neurosurgery (Baltimore), 37: 246-254, 1995. fragmentation. J. Cell Biol., 119: 493-501, 1992. 31. Rubio, M. P., von Deimling, A., Yandell, D. W., Wiestler, 0. D., 19. Gorcyza, W., Gong, J. P., and Darzynkiewicz, Z. Detection of DNA Guselba, J. F., and Louis, D. N. Accumulation of wild-type p53 in strand breaks in individual apoptotic cells by the in situ terminal de- human astrocytomas. Cancer Res., 53: 3465-3467, 1993.

Downloaded from clincancerres.aacrjournals.org on September 26, 2021. © 1998 American Association for Cancer Research. Apoptotic tumor cell death induced by estramustine in patients with malignant glioma.

C Vallbo, A T Bergenheim, P Bergström, et al.

Clin Cancer Res 1998;4:87-91.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/4/1/87

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/4/1/87. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.