Safety and Efficacy of High-Dose Chemotherapy withAutologous Stem Cell Transplantation for Patientswith Malignant Astrocytomas

Benjamin Chen, M.Sc.1

Tauseef Ahmed, M.D.2

Anney Mannancheril, M.D.2

Michael Gruber, M.D.3

Deborah L. Benzil, M.D.1

1 Department of Neurosurgery, New York MedicalCollege, Valhalla, New York.

2 Division of Oncology/Hematology, Department ofMedicine, New York Medical College, Valhalla,New York.

3 Division of Neuro-Oncology, Department of Neu-rology, New York University, New York, New York.

The authors thank Ms. Arlene Stolper Siman forexpert editorial guidance.

Address for reprints: Deborah L. Benzil, M.D., De-partment of Neurosurgery, Munger 3, New YorkMedical College, Valhalla, NY 10595; Fax: (914)594-3641; E-mail: deborah_benzil@nymc.edu

Received May 29, 2003; revision received Novem-ber 24, 2003; accepted February 23, 2004.

BACKGROUND. Malignant astrocytomas are among the most resistant tumors to

curative treatments. Mean survival without treatment is measured in weeks, and

even with maximal surgery and radiation, the mean reported survival is � 1 year.

The advent of supportive treatments and newer agents has resulted in benefits for

many patients with cancer. The authors investigated the safety and effect on

survival of a high-dose thiotepa and carboplatin regimen with autologous stem cell

transplantation (ASCT) in patients with malignant astrocytomas who were enrolled

in a prospective trial approved by an institutional review board (IRB).

METHODS. Twenty-one patients were enrolled in an IRB-approved, prospective

trial. After baseline testing was completed, patients underwent peripheral stem cell

mobilization with cyclophosphamide (4 g/m2) and etoposide (450 mg/m2) fol-

lowed by granulocyte– colony-stimulating factor (10 �g/kg). Peripheral stem cells

were harvested when leukocyte counts recovered. Patients received 2 cycles of

thiotepa (750 mg/m2) and carboplatin (1600 mg/m2) followed by infusion of the

preserved stem cells. The cycles were administered 6 –10 weeks apart. Primary

outcome measures were patient survival (Kaplan–Meier analysis) and treatment

toxicity (using National Cancer Institute common toxicity criteria).

RESULTS. Autologous stem cells were harvested effectively and transfused in all

patients. Kaplan–Meier survival analysis demonstrated a survival time of 34.3 � 5.5

months (range, 9 –94 months). Despite significant myelosuppression, only three

patients experienced Grade 4 complications and eight experienced Grade 3 com-

plications.

CONCLUSIONS. High-dose chemotherapy with thiotepa and carboplatin with con-

comitant ASCT was used safely to treat patients with malignant astrocytomas and

may provide a survival advantage. Cancer 2004;100:2201–7.

© 2004 American Cancer Society.

KEYWORDS: adjuvant chemotherapy, anaplastic astrocytoma, antineoplastic com-bined chemotherapy protocol, autologous transplantation, glioblastoma, hemato-poietic stem cells.

Malignant astrocytomas, including glioblastoma multiforme(GBM) and anaplastic astrocytoma (AA), have an incidence of 6.5

per 100,000 person years1 and are among the most resistant tumors tocurative treatment. These tumors have a high rate of local recurrencedespite surgical resection, adjuvant radiotherapy, and chemother-apy.2–5 Survival for patients after initial clinical presentation withoutany treatment is � 3 months. Standard management of patients withmalignant astrocytomas including surgical resection followed bypostoperative radiotherapy extends the median survival to 35– 48weeks6,7 with 2-year survival rates of only 10%.3 The use of adjuvant

2201

© 2004 American Cancer SocietyDOI 10.1002/cncr.20223Published online 19 April 2004 in Wiley InterScience (www.interscience.wiley.com).

chemotherapy to control disease progression or recur-rence has yielded modest results. An early study witha carmustine regimen showed an increase in the num-ber of individuals surviving at 18 months, although noimprovement was observed for median survival.6 Amore recent metaanalysis of 16 randomized clinicaltrials demonstrated a 10.1% (1-year) and an 8.6% (2-year) increase in survival for patients receiving adju-vant chemotherapy compared with patients receivingonly radiotherapy.8

The advent of supportive treatments such as he-matologic growth factors and autologous stem celltransplantation (ASCT) has made the administrationof high-dose chemotherapy (HDCT) possible, with po-tential benefits for patients with various types of can-cer. Retrospective studies suggest that tumor remis-sion or down-staging is more commonly achievedwith HDCT than with conventional therapy for pa-tients with certain cancers, including Hodgkin diseaseand breast and colorectal carcinomas.9 –11 Collabora-tive results from randomized prospective studies ofHDCT efficacy are emerging.12 Although to our knowl-edge little survival benefit of conventional postopera-tive chemotherapy has been demonstrated for pa-tients with malignant astrocytomas,13,14 the trueimpact of HDCT with ASCT on the survival of thesepatients has not been defined.

The blood– brain barrier prevents most conven-tional chemotherapeutic agents from reaching intrin-sic brain tumors. Agents formulated so that they nat-urally cross the blood– brain barrier might hold somepromise. We hypothesized that higher systemic con-centrations of agents with known effectiveness in vitromay result in more effective doses crossing the blood–brain barrier and reaching tumor cells. The purpose ofthe current study was to determine the safety andefficacy of a HDCT regimen with one drug known tohave high central nervous system (CNS) penetration(thiotepa) and one agent with known CNS activity(carboplatin) with ASCT in the treatment of malignantastrocytomas.

MATERIALS AND METHODSPatient PopulationThe trial of HDCT with ASCT received approval fromthe institutional review boards and began enrollingpatients in 1996. Important eligibility criteria (Table 1)included age � 18 years and histologic confirmation ofGBM or AA by a neuropathologist at the institutions inwhich the study was conducted. Disease could benewly diagnosed or recurrent. Previous radiation orsurgery did not preclude enrollment. Exclusion crite-ria were limited to recent treatment with a nitrosourea(� 6 weeks) or medical conditions that precluded tol-

erance of HDCT. All patients presenting who metstudy eligibility criteria were offered enrollment in thestudy. Informed consent was obtained from these in-dividuals. Twenty-one patients (17 males and 4 fe-males) with a mean age of 38 years were enrolled inthe study. The trial closed in 1999. Table 2 shows thegeneral characteristics of the 21 patients.

Trial ProtocolPatients underwent a physical examination; compre-hensive blood chemistry study; gadolinium-enhancedmagnetic resonance imaging of the brain (within 30days before the initiation of potentially myeloablativechemotherapy); chest X-ray; antibody titer analysis forcytomegalovirus, Epstein–Barr virus, human immuno-deficiency virus, and hepatitis; baseline multiple gatedacquisition study of ejection fraction; and pulmonaryfunction tests with carbon monoxide diffusion.

Peripheral stem cells were mobilized with 4 g/m2

of cyclophosphamide and 450 mg/m2 of etoposideover a 3-day period followed by 10 –15 �g/kg granulo-cyte– colony-stimulating factor (Amgen, ThousandOaks, CA) subcutaneously for 10 –14 days. Peripheralstem cells were collected when leukocyte counts re-covered. At least 4.0 � 106 CD34-positive cells/kg werecollected and cryopreserved with standard tech-niques.15

Chemotherapy was initiated after the completionof baseline studies. The first treatment day was desig-nated Day �6. Patients were given a daily intravenousdose of thiotepa (250 mg/m2) and carboplatin (533mg/m2) on Days �6, �5, and �4. Stem cells were

TABLE 1Criteria for Trial Enrollment

Eligibility criteria� 18 yrsHistologic confirmation of GBM or AANew or recurrent diseaseWith or without previous radiotherapy

Exclusion criteriaEuropean Cooperative Oncology Group performance score � 0Karnofsky performance score � 70Leukocyte count � 2 � 109/LPlatelet count � 1 � 1011/LCreatinine level � 2 mg/dLBilirubin level � 2 mg/dLEjection fraction � 40%Carbon dioxide diffusion � 40%Active infectionAnticoagulation intolerancePregnancyHIV infection

GBM: glioblastoma multiforme; AA: anaplastic astrocytoma; HIV: human immunodeficiency virus.

2202 CANCER May 15, 2004 / Volume 100 / Number 10

infused on Day 0. Two cycles of chemotherapy andASCT were administered 6 –10 weeks apart.

Supportive CareStandard supportive treatments during the duration ofprotocol included antiemetics, antibiotics, fungal pro-phylaxis, viral prophylaxis, and maintenance of plate-let counts � 20,000/mm3 via transfusion of irradiatedplatelets. Parenteral nutrition was administered ifneeded.

Outcome MeasuresPrimary outcome analysis was planned for patientsurvival and treatment-related toxicity. Patient toxicitywas monitored closely with appropriate laboratorystudies and was graded on the basis of the NationalCancer Institute common toxicity criteria, version 1.Patient survival was followed, and dates of death wererecorded at the time of final analysis (August 2002).Length of hospital stays and time to recovery of plate-let counts � 20,000/mm3 and leukocyte counts� 1,000/mm3 also were recorded.

Statistical AnalysisSurvival was measured from the date of histologicdiagnosis to the date of death or last contact. Kaplan–Meier survival distributions were calculated using Epi

Info 2000 (public domain software, available fromURL: http://www.cdc.gov/epiinfo). The Student t testfor unpaired data was used to compare the meansurvival times for the group that included (n � 21) andthe group that did not include (n � 19) the patientswith AA (null hypothesis that the difference betweenthe mean survival times of the groups is due to chancewith P � 0.05).

RESULTSEfficacy of ASCTAutologous peripheral stem cells (APSC) were effec-tively harvested and reinfused in all patients. Themean number of APSC reinfused during the firstcourse was 5.32 � 106 CD34-positive cells/kg (range,1.80 � 106–20.02 � 106 CD34-positive cells/kg). Themean number of APSC reinfused during the secondcourse was 5.34 � 106 CD34-positive cells/kg (range,2.20 � 106–21.68 � 106 CD34-positive cells/kg).

ToxicityDespite the expected significant myelosuppression,only two patients were reported to have Grade 3/4infectious complications (Table 3). Both respondedrapidly to antibiotics and recovered completely. Over-all, three patients experienced Grade 4 toxicity and

TABLE 2Patient Characteristics

Patientno. Gender

Age(yr)

Diagnosis Disease status attime oftreatment

Previoussurgery

PreviousradiationInitial Trial

1 M 50.9 — GBM New onset Macroscopic No2 F 28.7 — AMG New onset — No3 M 40.2 — GBM New onset Partial Yes4 M 21.3 PXA GBM Recurrent Macroscopic No5 M 59.5 — GBM New onset Partial No6 M 22.6 — GBM New onset Partial Yes7 M 41.8 GBM GBM Recurrent Macroscopic Yes8 M 36.9 A-II GBM Recurrent Macroscopic Yes9 M 45.3 — GBM New onset Partial Yes10 M 36.6 ODG GBM — Partial Yes11 M 33.3 — AMG New onset Partial No12 F 30.8 GBM GBM Recurrent Macroscopic Yes13 M 48.9 ODG GBM Recurrent Partial Yes14 M 54.9 — GBM New onset Macroscopic No15 M 19.9 — GBM New onset — —16 M 41.6 AMG GBM Recurrent Partial No17 M 28.4 — GBM New onset Biopsy Yes18 F 25.6 ODG GBM Recurrent Macroscopic No19 M 42.9 — GBM New onset Partial No20 F 50.0 GBM GBM Recurrent Macroscopic Yes21 M 36.2 GBM GBM Recurrent Macroscopic Yes

GBM: glioblastoma multiforme; ODG: oligodendroglioma; AMG: anaplastic mixed glioma; PXA: pleomorphic xanthoastrocytoma; A-II: astrocytoma Stage II; —: no information; M: male; F: female.

Astrocytomas and HDCT/Chen et al. 2203

eight patients experienced Grade 3 toxicity (Table 3).Grade 2 toxicity in nearly all the patients includednausea/emesis, which was managed easily with stan-dard supportive measures.

SurvivalNineteen of the 21 patients enrolled in the study diedof progressive disease. The remaining 2 patients werelost to follow-up after 6 months and their dates ofdeath could not be determined. Kaplan–Meier survivalanalysis was applied to the study population with 19uncensored and 2 censored observations. The meansurvival time (� the standard error of the mean [SEM])was 34.3 � 5.5 months (range, 9 –94 months) (Fig. 1).The Kaplan–Meier estimate for the probability of pro-gression-free survival was 0.47 � 0.11 at 24 months.When the 2 patients with newly diagnosed AA were

excluded from the study group, the mean survival time(� SEM) was 31.7 � 5.7 months (range, 9 –94 months)with a probability of progression-free survival of 0.41� 0.12 at 24 months. There was no significant differ-ence noted between the mean survival times of the

FIGURE 1. Kaplan–Meier estimates of survival after high-dose chemotherapy

with autologous stem cell transplantation.

TABLE 3Patient Toxicities, Survival, Length of Hospitalization, and Cell Recovery Times

Patient no.

Toxicities (grades 3/4)a

Survival(mos)

Hospitalization (days)b

Cell recovery time (days)

Plateletsc Leukocytesd

Cycle 1 Cycle 2 Cycle 1 Cycle 2 Cycle 1 Cycle 2 Cycle 1 Cycle 2

1 Mucositis (4) 27 19 27 17 18 17 182 46 � 10 � 13 15 15 15 153 Infection (4) 19 21 25 20 22 20 164 —e 16 — 15 — 16 —5 Nausea/emesis (3) 11 12 12 16 15 16 156 52 20 19 15 17 17 177 Mucositis (3) 46 � 7 — 14 — 15 —8 9 — � 7 — 17 — 169 Infection (3), pulmonary (4),

CNS (4)15 � 9 — 13 — 17 —

10 94 — — — — — —11 Diarrhea (3), epistaxis (3) 67 � 5 � 7 15 17 16 1712 80 — — — 18 — 1813 19 20 3 12 — 15 —14 Mucositis (3) 30 18 17 17 14 15 1515 — � 6 10 15 15 15 1516 30 10 13 18 21 19 1817 CNS (3) 22 14 16 14 17 17 1918 23 17 — 14 — 16 —19 18 — 20 — 20 — —20 CNS (3) 20 31 53 18 17 17 1521 CNS (3) 24 17 14 � 17 16 � 17 16Mean � SEM 34.3 � 5.5 14.8 � 1.6 17.1 � 3.1 15.6 � 0.5 17.3 � 0.6 16.5 � 0.3 16.4 � 0.4

CNS: central nervous system; SEM: standard error of the mean.a Grade according to National Cancer Institute common toxicity criteria, version 1. The patients who experienced mild Grade 1 and grade 2 toxicities are as follows: Cycle 1: nausea/emesis (n � 16), skin/mucositis

(n � 14), diarrhea (n � 13), infection (n � 9), CNS effects (n � 4), hemorrhage (n � 1), cough (n � 1), epistaxis (n � 2); Cycle 2: nausea/emesis (n � 14), skin/mucositis (n � 12), infection (n � 11), diarrhea (n

� 10), CNS effects (n � 5), pulmonary (n � 2), epistaxis (n � 1).b Cumulative length of hospital stays within treatment cycle. Not all patients were treated on an in-patient basis. Hospitalization may have occurred only for treatment-related complications or effects. The end date

of a hospital stay was the last date of entry in the study record.c Platelet count stabilized � 20,000/mm3 is considered recovery.d Leukocyte count stabilized � 1000/mm3 is considered recovery.e —: not available.

2204 CANCER May 15, 2004 / Volume 100 / Number 10

two groups (Student t test for unpaired data, P� 0.744).

Treatment ToleranceTreatment was well tolerated by 19 of the 21 patients.One patient developed an allergic reaction during thesecond cycle of treatment and was given an intrave-nous dose of thiotepa (750 mg/m2) alone in lieu ofthiotepa and carboplatin. No patients required care inan intensive care unit, respiratory support, or invasivemonitoring. There were no deaths due to treatment-related toxicity. Although not all patients were treatedon an inpatient basis, most patients required hospi-talization during each cycle of chemotherapy. Meanhospitalizations were only 14.8 � 1.6 days for Cycle 1and 17.1�3.1 days for Cycle 2. Platelet and leukocytecounts stabilized within 17 days for both cycles (Ta-ble 3).

Radiologic ResponseAlthough analysis of the radiographic response wasnot a primary goal of the current study, available im-aging studies demonstrated that tumor size decreasedin four patients, remained stable in two patients, andincreased in two patients during the course of treat-ment. Imaging data for the remaining patients wereincomplete.

DISCUSSIONDespite advances in understanding the biology of can-cer (for selected reviews, see references 16 –19), high-grade gliomas remain resistant to the best clinicalefforts. Certain treatments, including radiation,20 car-mustine,21 temozolomide,22–24 tamoxifen,25,26 and ste-reotactic radiosurgery,27 have resulted in small sur-vival advantages, particularly for patients age � 40years. We have shown that HDCT with thiotepa andcarboplatin with ASCT can be administered safely andappears to provide a better survival advantage com-pared with survivals reported in other published stud-ies.2–7,20 These results mirror those observed withother high-dose regimens28 –30 and contrast with re-sults observed with carmustine-containing regimens,for which treatment-related toxicity and mortalitywere observed.31–33 However, much of this survivaladvantage may be attributed to the young age of thestudy population (mean age, 38 years vs. 56 years forall patients with GBM34). Superior survivals occuramong younger patients with these tumors, an effectthat most likely is related to tumor biology and hostfactors. Unfortunately, these results suggest thathigher systemic concentrations of chemotherapeuticagents do not automatically result in improved effi-cacy beyond the blood– brain barrier.

Real benefits to patients with malignant astrocy-tomas may come only with truly innovative ap-proaches to treatments. The use of neoadjuvant che-motherapy before surgery or radiotherapy has showngreat promise in the treatment of a number of carci-nomas, including rectal,35–37 breast,38,39 and head andneck carcinomas.40,41 For these patients, neoadjuvantchemotherapy leads to easier and complete surgicalresection with improved survival and quality of life.For such an approach to be feasible, the chemother-apy regimen must be easily administered, well toler-ated, and not exacerbate clinical symptomatology. Inaddition, surgical intervention must remain possible,with maintenance of adequate hematologic countsand without imposing significant medical risk factors.Although the current study is not large enough to fullyevaluate the efficacy of the treatment protocol, thedrug regimen could meet the criteria for neoadjuvantuse given the rapid recovery of blood cells with rein-fusion of autologous stem cells. Recent studies suggestthat surgical resection of malignant astrocytomas mayprovide a survival benefit only if � 90% tumor resec-tion is possible.42

Neoadjuvant chemotherapy might also be consid-ered before radiotherapy, which usually followsshortly after diagnosis and surgery. In several studies,the efficacy of chemotherapy was diminished by pre-vious radiation.43– 45 Theoretically, hypoxia, the resultof fractionated radiation, could diminish chemother-apeutic mechanisms that rely on some degree of vas-cularization for drug delivery to the region involved.Because radiotherapy provides a survival advantage topatients with malignant astrocytomas, neoadjuvantchemotherapy would have to be delivered quickly af-ter surgery or diagnosis so as not to unduly delayradiotherapy.

With the supportive treatments now in use, HDCTregimens that have low toxicity are possible and, un-der the right circumstances, may be efficacious. HDCTwith thiotepa and carboplatin with concomitant ASCTcan be implemented safely for patients with malignantastrocytomas and may provide some survival advan-tage. This chemotherapy regimen or similar ones mayprove more useful as a neoadjuvant treatment.

REFERENCES1. Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics,

1999. CA Cancer J Clin. 1999;49:8 –31.2. Green SB, Byar DP, Walker MD, et al. Comparisons of car-

mustine, procarbazine, and high-dose methylprednisoloneas additions to surgery and radiotherapy for the treatment ofmalignant glioma. Cancer Treat Rep. 1983;67:121–132.

Astrocytomas and HDCT/Chen et al. 2205

3. Levin V, Leibel S, Gutin P. Neoplasms of the central nervoussystem. In: DeVita V Jr., Hellman S, Rosenberg S, editors.Cancer: principles and practice of oncology. Philadelphia:Lippincott-Raven, 1997:2022–2082.

4. Prados MD, Larson DA, Lamborn K, et al. Radiation therapyand hydroxyurea followed by the combination of 6-thiogua-nine and BCNU for the treatment of primary malignantbrain tumors. Int J Radiat Oncol Biol Phys 1998;40:57– 63.

5. Prados MD, Levin V. Biology and treatment of malignantglioma. Semin Oncol. 2000;27:1–10.

6. Walker MD, Green SB, Byar DR, et al. Randomized compar-isons of radiotherapy and nitrosoureas for the treatment ofmalignant glioma after surgery. N Engl J Med. 1980;303:1323–1329.

7. Young B, Oldfield EH, Markesbery WR, et al. Reoperation forglioblastoma. J Neurosurg. 1981;55:917–921.

8. Fine HA, Dear KB, Loeffler JS, Black PM, Canellos GP. Meta-analysis of radiation therapy with and without adjuvantchemotherapy for malignant gliomas in adults. Cancer.1993;71:2585–2597.

9. Czyz J, Hellmann A, Dziadziuszko R, et al. High-dose che-motherapy with autologous stem cell transplantation is aneffective treatment of primary refractory Hodgkin’s disease.Retrospective study of the Polish Lymphoma ResearchGroup. Bone Marrow Transplant. 2002;30:29 –34.

10. Huang E, McNeese MD, Strom EA, et al. Locoregional treat-ment outcomes for inoperable anthracycline-resistantbreast cancer. Int J Radiat Oncol Biol Phys. 2002;53:1225–1233.

11. Ragnhammar P, Hafstrom L, Nygren P, Glimelius B. A sys-tematic overview of chemotherapy effects in colorectal can-cer. Acta Oncol 2001;40:282–308.

12. Schmitz N, Pfistner B, Sextro M, et al. Aggressive conven-tional chemotherapy compared with high-dose chemother-apy with autologous haemopoietic stem-cell transplanta-tion for relapsed chemosensitive Hodgkin’s disease: arandomised trial. Lancet. 2002;359:2065–2071.

13. Galanis E, Buckner JC. Chemotherapy for high-grade glio-mas. Br J Cancer. 2000;82:1371–1380.

14. Huncharek M, Muscat J. Treatment of recurrent high gradeastrocytoma; results of a systematic review of 1,415 patients.Anticancer Res. 1998;18:1303–1311.

15. Stiff PJ, Koester AR, Weidner MK, Dvorak K, Fisher RI. Au-tologous bone marrow transplantation using unfractionatedcells cryopreserved in dimethylsulfoxide and hydroxyethylstarch without controlled-rate freezing. Blood. 1987;70:974 –978.

16. Aaronson SA. Growth factors and cancer. Science. 1991;254:1146 –1156.

17. Fang X. Lysophospholipid growth factors in the initiation,progression, metastases, and management of ovarian can-cer. Ann N Y Acad Sci. 2000;905:188 –208.

18. Favoni RE, deCupis A. The role of polypeptide growth fac-tors in human carcinomas: new targets for a novel pharma-cological approach. Pharm Rev. 2000;52:179 –206.

19. Westermark B, Nister M, Heldin CH. Growth factors andoncogenes in human malignant glioma. Neurol Clin. 1985;3:785–799.

20. Burger PC. Malignant astrocytic neoplasms: classification,pathological anatomy, and response to treatment. SeminOncol. 1986;13:16 –26.

21. Vitanovics D, Sipos L, Afra D. BCNU-DBD (dibromodulcitol)chemotherapy of recurrent supratentorial anaplastic astro-cytomas and glioblastomas. Neoplasma. 2002;49:342–345.

22. Friedman HS, Kerby T, Calvert H. Temozolomide and treat-ment of malignant glioma. Clin Cancer Res. 2000; 6:2585–2597.

23. Schwenka J, Ignoffo RJ. Temozolomide. A new option forhigh-grade astrocytomas. Cancer Pract. 2000;8:311–313.

24. Yung WA, Prados MD, Yaya-Tur R, et al. Multicenter phaseII trial of temozolomide in patients with anaplastic astrocy-toma or anaplastic oligoastrocytoma at first relapse. Te-modal Brain Tumor Group. J Clin Oncol. 1999;17:2762–2771.

25. Chamberlain MC, Kormanik PA. Salvage chemotherapy withtamoxifen for recurrent anaplastic astrocytomas. Arch Neu-rol. 1999;56:703–708.

26. Couldwell WT, Weiss MH, DeGiorgio CM, et al. Clinical andradiological response in a minority of patients with recur-rent malignant gliomas treated with high-dose tamoxifen.Neurosurgery. 1993;32:485– 490.

27. Kondziolka D, Flickinger JC, Bissonette DJ, Bozik M,Lunsford LD. Survival benefit of stereotactic radiosurgeryfor patients with malignant glial neoplasms. Neurosurgery.1997;41:776 –783.

28. Dunkel IJ, Finlay JL. High-dose chemotherapy with autolo-gous stem cell rescue for brain tumors. Crit Rev Oncol He-matol. 2002;41:197–204.

29. Finlay JL. The role of high-dose chemotherapy and stem cellrescue in the treatment of malignant brain tumors. BoneMarrow Transplant. 1996;18:S1–S5.

30. Abrey LE, Rosenblum MK, Papadopoulos E, Childs BH, Fin-lay JL. High dose chemotherapy with autologous stem cellrescue in adults with malignant brain tumors. J Neurooncol.1999;44:147–153.

31. Grovas AC, Boyett JM, Lindsley K, Rosenblum M, Yates AJ,Finlay JL. Regimen-related toxicity of myeloablative chemo-therapy with BCNU, thiotepa, and etoposide followed byautologous stem cell rescue for children with newly diag-nosed glioblastoma multiforme: report from the children’scancer group. Med Pediatr Oncol. 1999;33:83– 87.

32. Lassen U, Kristjansen PE, Wagner A, Kosteljanetz M,Poulsen HS. Treatment of newly diagnosed glioblastomamultiforme with carmustine, cisplatin and etoposide fol-lowed by radiotherapy. A phase II study. J Neurooncol. 1999;43:161–166.

33. Brandes AA, Palmisano V, Pasetto LM, Basso U, MonfardiniS. High-dose chemotherapy with bone marrow rescue forhigh-grade gliomas in adults. Cancer Invest. 2000;19:41– 48.

34. Greenberg MS. Handbook of neurosurgery, 3rd ed. Lake-land, FL: Greenberg Graphics, 1994:612.

35. Kim JS, Kim JS, Cho MJ, Song KS, Yoon WH. Preoperativechemoradiation using oral capecitabine in locally advancedrectal cancer. Int J Radiat Oncol Biol Phys. 2002;54:403– 408.

36. Glimelius B. Chemoradiotherapy for rectal cancer—is therean optimal combination? Ann Oncol. 2001;12:1039 –1045.

37. Schaffer M, Thoma M, Wilkowski R, Schaffer P, Duhmke E.Radio-chemotherapy as a preoperative treatment for ad-vanced rectal cancer. Evaluation of down-staging and mor-bidity. Onkologie. 2002;25:352–356.

38. Hortobagyi GN. Multidisciplinary management of advancedprimary and metastatic breast cancer. Cancer. 1994;74:S416 –S423.

39. Valero V, Buzdar AU, McNeese M, Singletary E, HortobagyiGN. Primary chemotherapy in the treatment of breast can-cer: the University of Texas M.D. Anderson Cancer Centerexperience. Clin Breast Cancer. 2002;3:S63–S68.

2206 CANCER May 15, 2004 / Volume 100 / Number 10

40. Basu S, Khanra M, Dash B, Majumdar J, Biswas J, ChaudhuriP. The role of neoadjuvant and adjuvant chemotherapyregimens consisting of different combinations of drugs inthe treatment of advanced oral cancer. Med Oncol. 1999;16:199 –203.

41. Grau JJ, Domingo J, Blanch JL, et al. Multidisciplinary ap-proach in advanced cancer of the oral cavity: outcome withneoadjuvant chemotherapy according to intention-to-treatlocal therapy. A phase II study. Oncology. 2002;63:338 –345.

42. Lacroix M, Abi-Said D, Fourney DR, et al. A multivariateanalysis of 416 patients with glioblastoma multiforme: prog-

nosis, extent of resection, and survival. J Neurosurg. 2001;95:190 –198.

43. Teicher BA. Hypoxia and drug resistance. Cancer MetastasisRev. 1994;13:139 –168.

44. Yount GL, Haas-Kogan DA, Levine KS, Aldape KD, Israel MA.Ionizing radiation inhibits chemotherapy-induced apopto-sis in cultured glioma cells: implications for combined mo-dality therapy. Cancer Res. 1998;58:3819 –3825.

45. Yu JL, Coomber BL, Kerber RS. A paradigm for therapy-induced microenvironmental changes in solid tumors lead-ing to drug resistance. Differentiation. 2002;70:599 – 609.

Astrocytomas and HDCT/Chen et al. 2207