Impact of combination therapy with repeat surgery and temozolomide for recurrent or progressive glioblastoma multiforme: a prospective trial

Surgical Neurology 68 (2007) 250 – 254 www.surgicalneurology-online.com

Neoplasm

Impact of combination therapy with repeat surgery and temozolomide for recurrent or progressive glioblastoma multiforme: a prospective trial Mizuhiko Terasaki, MD, PhDa,4, Etsuyo Ogo, MD, PhDb, Shintaro Fukushima, MDa, Kiyohiko Sakata, MD, PhDa, Naohisa Miyagi, MD, PhDa, Toshi Abe, MD, PhDb, Minoru Shigemori, MD, PhDa Departments of aNeurosurgery and bRadiology, Kurume University School of Medicine, Kurume, Fukuoka 830-0011, Japan Received 22 September 2006; accepted 1 November 2006

Abstract

Background: The purpose of this study is to evaluate the feasibility of repeat surgery combination with postoperative TMZ for adults with recurrent or progressive GBMs. Methods: Of 35 patients who had diagnoses of GBM between 2002 and 2005, 7 (20%) underwent second surgeries and TMZ for recurrent or progressive disease. We examined the case histories of these 7 patients and determined the location of tumor and extent of their surgical procedures. Using the KPS, we assessed each patient’s neurologic state before and after initial and repeat surgery. We calculated survival times from time of initial surgery and compared actual survival with statistically predicted survival times for each patient. Results: Median survival rates were higher than predicted: The statistical risk estimate for median survival time after repeat surgery for these patients was 9 months; the actual survival time from initial operation until time of death averaged 15.1 months. The neurologic status before and after second surgery also averaged 76 points on KPS. Survival time did not depend on removing the entire tumor at initial and second surgery. Conclusion: Combination with repeat surgery and TMZ improved overall survival of these GBM patients. D 2007 Elsevier Inc. All rights reserved.

Keywords:

Glioblastoma; Repeat surgery; Survival; Temozolomide

1. Introduction The prognosis for adults with recurrent or progressive GBMs is dismal [2-4,10,15,16]. Treatment with conventional chemotherapy has not proven effective [7], and more innovative approaches such as stereotactic radiosurgery have resulted in only modest success [11]. Other approaches, including repeat surgery may be a matter of debate. There have been reports concerning the use of repeat

Abbreviations: ACNU, Nimustine; CNS, central nervous system; CSF, cerebrospinal fluid; GBM, glioblastomas; GTR, gross total resection; HBO, hyperbaric oxygenation; KPS, Karnofsky performance status; PR, partial response; RT, radiotherapy; STR, subtotal resection; TMZ, temozolomide; T2WI, T2-weighted image. 4 Corresponding author. Tel.: +81 942 35 3311; fax: +81 942 38 8179. E-mail address: [email protected] (M. Terasaki). 0090-3019/$ – see front matter D 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.surneu.2006.11.042

surgery to debulk recurrent tumors [8]. Although this is a favorable approach for some patients, the surgery itself carries a significant risk of morbidity. In their series of 36 adults with initial surgery, the survival benefit although significant was only 3 months. The question bShould a recurring GBM be surgically removed, and if so, when?Q is of great interest and debate among neurosurgeons. Temozolomide is a recently developed alkylating agent that has proven activity against adult CNS malignancies [5,6,13,20]. Despite initial enthusiasm about TMZ for adults with brain tumors, in a recent phase II study of this agent for adults with recurrent CNS malignancies, only 5% of adults with relapsed or progressive GBMs responded to therapy [1,20]. We postulated that combination therapy with repeat surgery and TMZ might result in prolonged survival times. This study investigates the experience at our institution with

M. Terasaki et al. / Surgical Neurology 68 (2007) 250 – 254

251

2. Materials and methods

and vincristine were given. Two courses of chemotherapy were given at 6-week intervals synchronously with RT-HBO. Granulocyte colony-stimulating factor was given when the granulocyte counts in the peripheral blood fell to less than 1000/mL.

2.1. Eligibility

2.3. Induction of repeat surgery

Adults with recurrent or progressive disease at confirmatory prediagnosed supratentorial GBMs were considered eligible for this protocol. Patients with different tumor site and histologic diagnoses with mixed glial components were excluded from this study. Evidence of neuraxis or extraneural dissemination was also excluded. Patients could have previously received either irradiation and/or chemotherapy. Of 35 patients who had an initial diagnosis of histologically confirmed supratentorial GBM between 2002 and 2005, 7 consented patients with recurrent or progressive GBM were eligible and treated at Kurume University, Fukuoka, Japan. All patients gave informed consent. We retrospectively compiled data from these 7 patients. We used the date of initial histologic confirmation of diagnosis as the date of disease onset.

Patients with progressive disease during and/or after the initial treatment, or recurrent disease after the completion of the initial treatment, underwent a second operation. If residual tumor responded to the initial adjuvant treatment, we delivered the same chemotherapeutic regimen as maintenance chemotherapy. After repeat surgery, patients received TMZ 150 to 200 mg/m2 a day from days 1 to 5. Cycles were repeated every 28 days.

repeat surgery and TMZ for the treatment of recurrent or progressive GBM.

2.4. Pathologic review Tumor pathology slides were centrally reviewed at Kurume University and classified according to the World Health Organization criteria. 2.5. Patient monitoring/toxicity criteria Complete blood cell count and biochemical studies were to be obtained frequently throughout TMZ chemotherapy. The National Cancer Institution Toxicity Criteria was used to grade toxicity. At prescribed times during and after chemotherapy, the disease status of all patients was monitored with appropriate neurologic examinations, CSF cytologic examinations, and neuroimaging studies to assess the effect of TMZ chemotherapy on postoperative residual tumors and T2WI lesions. Criteria for response/relapse by neuroradiologic studies were defined as follows: CR, complete resolution of tumor on enhanced magnetic resonance imaging scan; PR, a greater than 50% reduction in the product of the greatest tumor diameter and its perpendicular diameter as measured on an enhanced magnetic resonance imaging scan; minor response, 25% to 50% reduction in the product of the greatest

2.2. Initial treatment regimen All patients had initial surgery. Six also had adjuvant therapy with a combination of alkylating-based chemotherapy and RT synchronized with HBO. Radiotherapy consisted of 60 Gy in 30 fractions to the tumor area, determined from T2WI magnetic resonance images, and an additional 2-cm margin. Patients received treatment with HBO in a single-place hyperbaric chamber. Each irradiation was administered within 15 minutes after decompression [12]. Chemotherapy, consisting of procarbazine, ACNU, cepharanthine, and vincristine, was administered for 12 consecutive days (one course). ACNU was administered as soon as possible after HBO exposure to maximize drug delivery [14]. Two hours after the ACNU induction, cepharanthine Table 1 Patient data Patient no.

Age (y)/ sex

Tumor site

Tumor location

Initial surgery

Adjuvant therapy

1

66/F

B

B

STR

None

2

50/M

Fr

R

GTR

3

54/M

T

L

GTR

4

72/M

T

L

GTR

5

18/F

Fr

R

STR

6

63/F

Fr

L

STR

7

39/F

Fr

L

GTR

RT/ACNU/ VCR RT/ACNU/ VCR RT/ACNU/ VCR RT/ACNU/ VCR RT/ACNU/ VCR RT/ACNU/ VCR

Initial KPS

KPS after first surgery

Time from initial surgery (mo)

KPS before second surgery

Second surgery

KPS after second surgery

PFS

Outcome

DOD 16 mo DOD 17 mo DOD 16 mo DOD 13 mo DF 20+ mo DOD 12 mo AWD 12+ mo

80

80

4

80

STR

80

12

100

100

12

100

GTR

100

3

80

80

1

70

GTR

70

6

60

60

5

50

GTR

50

3

100

100

11

100

GTR

100

9+

30

30

7

30

STR

30

1

100

100

8

100

GTR

100

4+

PFS indicates progression-free survival; F, female; B, bifocal; DOD, died of disease; M, male; Fr, frontal; R, right; ACNU/VCR, procarbazine, nimustine, cepharanthine, and vincristine; T, temporal; L, left; DF, disease free; AWD, alive with disease.

252

M. Terasaki et al. / Surgical Neurology 68 (2007) 250 – 254

tumor diameter and its perpendicular diameter; stable disease, a less than 25% decrease in the product of the perpendicular diameters described; progressive disease, a greater than 25% increase in the tumor size, or appearance of tumor in a previously uninvolved area. 2.6. Informed consent Signed informed consent was obtained for each adult or legal spouse. Potential risks and benefits, therapeutic alternatives such as conventional chemotherapy and irradiation, and supportive care issues were discussed. Protocols were approved by each local institutional review board. 2.7. Analysis We calculated survival time as the time from initial surgery until the end of follow-up or death. We defined the surgery for removal of GBM as total resection when the tumor was completely removed or subtotal resection when residual tumor remained. We assessed each patient’s neurologic state with the standardized KPS index before and after each surgical procedure. Because this was a retrospective study, the parameters we used for the KPS neurologic deficiencies were associated with tumors, age, sex, and localization of the GBM. A high KPS score indicated high functional and neurologic ability. We used the Kaplan-Meier method to determine the statistical risk estimate for survival time after repeat surgery. We then calculated survival times from time of initial surgery and compared actual survival with statistically predicted survival times for each patient. 3. Results 3.1. Patient demographics Table 1 summarizes the patient data. The median age of the 4 female and 3 male patients at the time of initial treatment was 52 years (range, 18-72 years). Tumors were located mainly in the frontal region. Six of 7 patients received prior surgery, chemotherapy, and RT, and 1 had prior initial surgery only. The interval between first and second surgery was less than 6 months in 43% of cases (3 patients). The median time of repeat surgery from the initial surgery was 6.9 months (range, 1-12 months). 3.2. Surgical results and responses Total resection was achieved in 43% of patients during the first operation and 71% during repeat surgery. Table 2 Table 2 Types of surgery and survival rate First/second operation, type of surgery

No. of patients

Average survival time (mo)

Total/total Subtotal/total Subtotal/subtotal

4 1 2

14.5 20 14

Fig. 1. Statistic calculation (Kaplan-Meier) of survival time.

shows the correlation between survival time and the types of resection (total and subtotal) at first and second surgery. When we compared first and second surgeries and postoperative survival times, we found that (1) total resection at initial surgery and second surgery did not substantially increase survival time, (2) the survival times for patients with subtotal resections at initial and second surgery were similar to those with total resections, and (3) average survival time did not depend on removing the entire tumor at either surgery. Two patients had postoperative complications at repeat surgery. Tumor location affected complications. Resection of the 4 GBMs in the frontal region resulted in no new neurologic deficits; however, removal of left-sided temporal GBMs produced sensory aphasia postoperatively. Of the 2 patients with subtotally resected GBM, none of patients achieved response but 1 patient demonstrated stable disease for 12 months. Of the 5 patients with grossly resected GBM, 1 patient (20%) achieved PR and 2 patients achieved stable disease of T2WI lesions by the end of 6 cycles of TMZ chemotherapy, for a total objective response (CR + PR) rate of 14%. One patient went on to receive TMZ chemotherapy. 3.3. Outcomes Before their initial operation, the neurologic status of patients on KPS averaged 79 points, and 5 patients (71%) had 80 points or more. After initial surgery, the average neurologic status decreased to 76 points. Patients were unable to carry on normal activity or do active work. The neurologic status before and after second surgery also averaged 76 points on KPS. Five patients died, and 2 patients remained disease free or alive with disease. The median follow-up was 15.1 months. The time-to-relapse rates ranged from 1 to 12 months after repeat surgery. The median from the Kaplan-Meier statistical risk estimate for survival time after repeat surgery was 9 months; the actual average time from

M. Terasaki et al. / Surgical Neurology 68 (2007) 250 – 254

initial surgery until time of death or end of follow-up was 15.1 months (Fig. 1). 3.4. Toxicity Toxicity was well tolerated in the 38 cycles of TMZ therapy administered. One patient had stomatitis after her eighth cycle. Two patients had grade 2 neutropenia. The remaining 5 had grade 0 or 1 neutropenia. One reported nausea or vomiting, but no patient was hospitalized for complications of TMZ chemotherapy.

4. Discussion Our results suggest that combination with second surgery and postoperative TMZ improves survival. However, our most significant finding concerns survival time and the degree of tumor resection at initial and repeat surgery; the total resection of recurrent or progressive tumor at repeat surgery with TMZ does not appear to increase survival time. After total resection at initial and second surgery, patients survived for an average of 14.5 months. Patients with subtotal resection at first and second surgery survived 14 months. This raises the question whether it would be more beneficial for the patient to undergo subtotal surgery the second time with TMZ. Because no additional literature is available on this issue, the influence of total resection at repeat surgery still remains uncertain and should be further examined. We delivered combination chemotherapy with nimustine as an initial treatment in this series. Temozolomide have been widely accepted as routine treatment and evaluated by some authors to determine both the rate and duration of response [9,17,18]. In a series with TMZ alone, the progression-free survival rate was 54% at 6 months [18]. In that series, a median OS was 14.6 months with 20% of grades 3 and 4 hematopoietic toxicity. Our results in this series of patients with newly diagnosed GBMs are consistent with the findings of previously published reports with chemotherapy and radiation therapy. Given the small numbers of patients treated in this series, the variability profile in eligibility, no further conclusions can be drawn concerning the superiority of one regimen over another. No studies on the efficacy of nimustine in the adults have been carried out. The drug regimen was too toxic with 47% having grade 3 and 4 neutropenia (data not shown); one had treatment terminated because of hematologic toxicity. In total, 83% of the patients who began the regimen of nimustine completed all the planned treatment. The progression-free survival rate in the patients treated with nimustine was 67% at 6 months. In single-agent trials of TMZ in adults with recurrent brain tumors, results were disappointing. Yung et al [20] reported only 5% of adults with recurrent GBMs had responses to TMZ. We note that several of the patients in our series had received alkylating agents before their recurrences. It is also possible that 6 of our patients had

253

prior chemotherapy and may have had more chemoresistant tumors. Furthermore, 86% of our patients had prior treatment with nitrosoureas. Because the mechanism of resistance to nitrosoureas and TMZ is similar, patients who progress after nitrosoureas may be less likely to respond to TMZ. Thus, these patients may have been a more chemoresistant group because of their lack of response to the nitrosoureas. Ultimately, adjuvant therapy also produces adverse effects and is as controversial as with second surgery. Nonetheless, TMZ alone was less toxic and may have caused tumor control in our patients. Despite this prior treatment, most patients had only grade 0 to 1 hematotoxicity. The only other toxicity reported was grade 2 stomatitis in one patient. The patients participating in our study received TMZ and survived, on average, 15 months. Their subsequent responses to the combination of repeat surgery and TMZ suggests greater activity of TMZ in combination rather than of either one alone. A further important parameter for survival prognosis was the neurologic status of patients on the KPS. After second surgery, patients had an average KPS of 76. They were unable to carry on normal activity or do active work. Taveras et al stress the uncertainty of a neurologic statistic with too many variables beginning with the subjective classification of the patients with individual values [19]. An objective total picture must be formed in assessing a patient for second surgery. One approach could be reducing the amount of resection at second surgery. This study suggests that total surgical resection in the repeat operation has little influence on survival time when combined with TMZ. Another approach could be the further improvement of adjuvant therapy. The decision about which therapy should be carried out has to be orientated to an interdisciplinary bquality-of-life conceptQ that takes into account the individual and situation-specific circumstances.

5. Conclusions Our experience in a series of 7 adults with recurrent or progressive GBMs substantiates the potential efficacy of combination therapy with repeat surgery and TMZ in at least prolonging the survival for most of the adults with such tumors. Selection bias may have played a role in deciding which adults had second surgeries. Also, we did not use TMZ as an initial treatment in this series. Furthermore, patients received repeat surgery and postoperative TMZ at varying times, making it difficult to define the role and the optimal timing of repeat surgery. Despite these limitations, this series demonstrated that combination therapy with repeat surgery and TMZ is a promising approach for recurring or progressing GBMs. We are now going to evaluate these combination effects when TMZ is used as routine treatment. The extended survival beyond the statistical prediction makes this approach attractive for further study.

254

M. Terasaki et al. / Surgical Neurology 68 (2007) 250 – 254

References [1] Brada M, Hoang-Xuan K, Rampling R, Dietrich PY, Dirix LY, Macdonald D, Heimans JJ, Zonnenberg BA, Bravo-Marques JM, Henriksson R, Stupp R, Yue N, Bruner J, Dugan M, Rao S, Zaknoen S. Multicenter phase II trial of temozolomide in patients with glioblastoma multiforme at first relapse. Ann Oncol 2001;12:259 - 66. [2] Burger PC, Green SB. Patient age, histologic features, and length of survival in patients with glioblastoma multiforme. Cancer 1987;59: 1617 - 25. [3] Burger PC, Vogel FS, Green SB, Strike TA. Glioblastoma multiforme and anaplastic astrocytoma. Pathologic criteria and prognostic implications. Cancer 1985;56:1106 - 11. [4] Busy PC. Glioblastomas. Surg Neurol 1985;24:589 - 90. [5] Estlin EJ, Lashford L, Ablett S, Price L, Gowing R, Gholkar A, Kohler J, Lewis IJ, Morland B, Pinkerton CR, Stevens MC, Mott M, Stevens R, Newell DR, Walker D, McDowell H, Reidenberg P, Statkevich P, Marco A, Batra V, Dugan M, Pearson AD. Phase 1 study of temozolomide in paediatric patients with advanced cancer. United Kingdom Children’s Cancer Study Group. Br J Cancer 1998; 78:652 - 61. [6] Gilbert MR, Friedman HS, Kuttesch JF, Prados MD, Olson JJ, Reaman GH, Zaknoen SL. A phase II study of temozolomide in patients with newly diagnosed supratentorial malignant glioma before radiation therapy. Neurooncology 2002;4:261 - 7. [7] Gonzalez J, Gilbert MR. Treatment of astrocytoma. Curr Opin Neurol 2005;18:632 - 8. [8] Guyotat J, Signorelli F, Frappaz D, Madarassy G, Ricci AC, Bret P. Is reoperation for recurrence of glioblastoma justified? Oncol Rep 2000;7:899 - 904. [9] Hegi ME, Stupp R. Correlative studies in neuro-oncology trials: should they influence treatment? Curr Oncol Rep 2006;8:54 - 7. [10] Heppner F. The glioblastoma multiforme: a lifelong challenge to the neurosurgeon. Neurochirurgia 1986;29:9 - 14. [11] Hsieh PC, Chandler JP, Bhangoo S, Panagiotopoulos K, Kalapurakal JA, Marymont MH, Cozzens JW, Levy RM, Salehi S. Adjuvant

[12]

[13]

[14]

[15] [16] [17]

[18]

[19] [20]

gamma knife stereotactic radiosurgery at the time of tumor progression potentially improves survival for patients with glioblastoma multiforme. Neurosurgery 2005;57:684 - 92. Kohshi K, Kinoshita Y, Terashima H, Konda N, Yokota A, Soejima T. Radiotherapy after hyperbaric oxygenation for malignant gliomas: a pilot study. J Cancer Res Clin Oncol 1996;122:676 - 8. Nicholson HS, Krailo M, Ames MM, Seibel NL, Reid JM, Liu-Mares W, Vezina LG, Ettinger AG, Reaman GH. Phase I study of temozolomide in children and adolescents with recurrent solid tumors: a report from the Children’s Cancer Group. J Clin Oncol 1998;16: 3037 - 43. Ogawa K, Yoshii Y, Inoue O, Toita T, Saito A, Kakinohana Y, Adachi G, Ishikawa Y, Kin S, Murayama S. Prospective trial of radiotherapy after hyperbaric oxygenation with chemotherapy for high-grade gliomas. Radiother Oncol 2003;67:63 - 7. Ransohoff J, Lieberman A. Surgical therapy of primary malignant brain tumors. Clin Neurosurg 1978;25:403 - 11. Salcman M. Survival in glioblastoma: historical perspective. Neurosurgery 1980;7:435 - 9. Stupp R, Hegi ME, van den Bent MJ, Mason WP, Weller M, Mirimanoff RO, Cairncross JG, European Organisation for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups, National Cancer Institute of Canada Clinical Trials Group. Changing paradigms—an update on the multidisciplinary management of malignant glioma. Oncologist 2006;11:165 - 80. Stupp R, van den Bent MJ, Hegi ME. Optimal role of temozolomide in the treatment of malignant gliomas. Curr Neurol Neurosci Rep 2005;5:198 - 206. Taveras JM, Thompson Jr HG, Pool JL. Should we treat glioblastoma multiforme? AJNR Am J Neuroradiol 1962;87:473 - 9. Yung WK, Albright RE, Olson J, Fredericks R, Fink K, Prados MD, Brada M, Spence A, Hohl RJ, Shapiro W, Glantz M, Greenberg H, Selker RG, Vick NA, Rampling R, Friedman H, Phillips P, Bruner J, Yue N, Osoba D, Zaknoen S, Levin VA. A phase II study of temozolomide vs. procarbazine in patients with glioblastoma multiforme at first relapse. Br J Cancer 2000;83:588 - 93.