Journal of Neuro-Oncology

, Volume 107, Issue 2, pp 395–405

Hypofractionated radiotherapy with or without concurrent temozolomide in elderly patients with glioblastoma multiforme: a review of ten-year single institutional experience

Authors

  • Jeffrey Q. Cao
    • Department of Oncology, London Regional Cancer Program, London Health Sciences CentreUniversity of Western Ontario
    • Department of Oncology, London Regional Cancer Program, London Health Sciences CentreUniversity of Western Ontario
  • Glenn S. Bauman
    • Department of Oncology, London Regional Cancer Program, London Health Sciences CentreUniversity of Western Ontario
  • Joseph F. Megyesi
    • Department of Neurosurgery, London Health Sciences CentreUniversity of Western Ontario
  • Christopher J. Watling
    • Department of Neurology, London Health Sciences CentreUniversity of Western Ontario
  • David R. Macdonald
    • Department of Oncology, London Regional Cancer Program, London Health Sciences CentreUniversity of Western Ontario
    • Department of Neurology, London Health Sciences CentreUniversity of Western Ontario
Clinical Study - Patient Study

DOI: 10.1007/s11060-011-0766-3

Cite this article as:
Cao, J.Q., Fisher, B.J., Bauman, G.S. et al. J Neurooncol (2012) 107: 395. doi:10.1007/s11060-011-0766-3

Abstract

The landmark Stupp study demonstrated a survival advantage with concomitant and adjuvant temozolomide (TMZ) with standard radiotherapy (RT) in glioblastoma multiforme (GBM) patients but excluded those older than 70 years. The prospective Roa study of older GBM patients treated with hypofractionated 3-week course RT demonstrated equivalence to standard 6-week course RT. Taken together, these trials suggest hypofractionated RT with TMZ may be a reasonable treatment option for elderly GBM patients. We conducted a retrospective review of GBM patients (age ≥60 years) treated with hypofractionated RT and temozolomide at our institution between 2000 and 2010. We identified 112 patients who received hypofractionated RT, with 57 receiving concurrent and adjuvant TMZ and 55 without concurrent chemotherapy. Of the 55 patients who received hypofractionated RT alone initially, 24 subsequently received TMZ as salvage treatment at time of progression. Among the concurrent RT + TMZ patients, mean age was 70 years (range 60–86), median KPS was 80 (range 30–100) and 24/57 (42%) received prior debulking surgery. Median overall survival (OS) among the RT + TMZ patients was 6.9 months (95% CI, 4.5–8.6). Patients without concurrent chemotherapy were similar in demographics (age, sex, corticosteroid use, KPS) except 34/55 (62%) were debulked (P-value 0.045.) Median OS was 9.3 months (95% CI, 5.9–11.8) (P-value 0.351). Sub-group analysis revealed patients treated with initial hypofractionated radiation with salvage TMZ had increased median OS of 13.3 months (95% CI, 9.9–19.3) (P-value 0.012). Our results suggest concurrent and adjuvant TMZ does not confer a survival benefit in elderly GBM patients. A sequential approach may be a more effective and efficient strategy by selecting responding patients who may benefit most from subsequent salvage chemotherapy.

Keywords

ElderlyGlioblastoma multiformeHypofractionatedRadiotherapyTemozolomide

Introduction

Glioblastoma multiforme (GBM) is the most aggressive malignant primary brain tumour and is the most frequently diagnosed glioma in adults [1]. It has been well established that age and performance status are significant prognostic indicators in patients with malignant glioma [24]. The frequency of GBM increases with age, with over 50% of glioblastomas occurring in individuals over age 70 [5]. Despite these demographics many clinical trials exclude older patients due to concerns regarding reduced functional reserve, poorer performance status and additional medical co-morbidities. Furthermore, those studies that do address treatment of these patients are inconsistent in their definition of “elderly”, with reports mainly including patients whose age ranges from 60 to 70 years. Therefore, there is currently no commonly accepted standard of care for the treatment of GBM in the elderly population [6]. The treatment options range from best supportive care (BSC) alone, surgical resection, radiation therapy (RT) alone, chemotherapy alone, or a combination of chemotherapy and radiation therapy, following surgical resection.

The landmark study by Stupp et al. [7] demonstrated a survival advantage for concomitant and adjuvant temozolomide (TMZ) chemotherapy added to a standard course of RT (60 Gy in 30 daily fractions over 6 weeks) compared to RT alone. The overall survival benefit of this aggressive treatment, however, was attenuated in older or poor performance status patients [8]. Additionally, this study excluded patients older than 70 years of age. In another prospective randomized control trial reported by Roa et al. [9] older GBM patients (age >60 years) treated with a shorter hypofractionated course of RT (40 Gy in 15 daily fractions over 3 weeks) demonstrated similar survival and palliative benefit compared to a standard 6-week course (overall survival 5.6 months for the 3-week group and 5.1 months for the 6-week group (hazard ratio, 0.89; 95% CI, 0.59–1.36; P = 0.57). Taken together, these two prominent trials suggested that a hypofractionated course of RT with concurrent and adjuvant TMZ may be a reasonable treatment option for elderly patients newly diagnosed with GBM. At our institution, we started offering hypofractionated radiotherapy as an option to patients in the early part of the decade. When the results of the Stupp trial became available, we began combining this approach with concurrent and adjuvant TMZ. We conducted this retrospective review to help determine if the strategy of upfront concurrent TMZ and hypofractionated radiotherapy was more effective than a policy of hypofractionated radiotherapy alone with salvage TMZ reserved for progression.

Methods

Retrospective review

After obtaining approval from our institutional Research Ethics Board, we conducted a retrospective review of elderly patients newly diagnosed with GBM between January 2000 and December 2009 and treated at the London Regional Cancer Program (London Health Sciences Centre, London, Ontario, Canada). The medical records were screened to include patients aged 60 years or older at time of diagnosis who received hypofractionated RT (40 Gy in 15 daily fractions over 3 weeks). Among this cohort we identified patients who received combined RT and TMZ chemotherapy given concurrently and adjuvantly. We also identified those patients in whom TMZ was not given or was delivered at the time of tumor progression post radiation. Data extracted from the retrieved charts and entered into a database included patient age, sex, symptoms prior to diagnosis, symptom duration prior to diagnosis, prior malignancy, pre-treatment imaging characteristics, tumour location, steroid use, date of diagnosis, extent of surgery, Karnofsky Performance Status (KPS) scores, radiation treatment dates, doses and technique, clinical response post-treatment, chemotherapy treatment dates and cycles, date of either clinical or radiographic progression, location of failure, salvage treatment for progression, and date of last known status.

Treatment details

Radiotherapy was administered by a hypofractionated schedule to a total dosage of 40 Gy in 15 fractions, at a dose of 2.67 Gy per fraction given once daily, five days per week, over a period of three weeks. Patients were immobilized in custom thermoplastic shells for simulation and treatment with dedicated computed tomography (CT) planning utilized with three dimensional-conformal radiation therapy delivery in all cases.

All patients receiving combined therapy were treated with concurrent chemotherapy consisting of temozolomide at a dose of 75 mg per square metre per day, given 7 days per week from the first day of radiotherapy until the last day of radiotherapy. Oral trimethoprim–sulfamethoxazole was prescribed three times per week during concurrent chemoradiation to mitigate the risk of Pneumocystis carinii pneumonia due to temozolomide-induced lymphocytopenia. Antiemetic prophylaxis with prochlorperazine and/or a 5-hydroxytryptamine-3 (5-HT3) antagonist was typically prescribed prior to concurrent and adjuvant TMZ chemotherapy. Following a 4-week break, concurrent TMZ patients subsequently received adjuvant temozolomide according to the standard 5-day regimen every 28 days for a planned six cycles. The initial dose was 150 mg per square metre for the first cycle and then increased to 200 mg per square metre with the second cycle, provided that toxicity was acceptable. Patients receiving salvage TMZ received the same regimen as the adjuvant phase of the chemoradiation patients.

Surveillance and follow-up

Tumor progression on follow-up imaging was defined according to the Macdonald criteria as an increase in tumor size by 25 percent or the appearance of new lesions [10]. Individuals presenting with interval clinical deterioration suggestive of tumor progression had imaging to confirm tumor progression. At the time of confirmed tumor progression, patients were treated at the discretion of the treating neuro-oncologist and the types of second-line therapy were recorded.

Statistical analysis

The date of the stereotactic biopsy or surgical resection providing a definitive histological diagnosis of GBM pathology (World Health Organization (WHO) Grade IV astrocytoma) was used as the date of definitive diagnosis. Clinical and/or radiographic progression-free survival was defined as the interval from date of diagnosis to date of clinical and/or radiographic progression (whichever was earlier). Overall survival was defined as the interval between date of diagnosis to date of death. Progression-free survival and overall survival were estimated using the Kaplan–Meier method. Two-sided log-rank tests statistics were used to assess the survival differences of the two treatment arms. Cox proportional hazards regression was used to assess between treatment arm differences adjusting for the influence of covariates.

Results

Patient characteristics

Between January 2000 and December 2009, we identified 112 elderly (age ≥60 years) GBM patients who received hypofractionated RT at our institution. Of these patients, 57 received hypofractionated RT plus concurrent and adjuvant TMZ (“concurrent”) and 55 elderly GBM patients who received the same short-course RT alone. Of those 55 patients treated without concurrent chemotherapy, 24 were treated subsequently with TMZ administered at the time of clinical or radiographic progression (“salvage”).

The median follow-up for the entire cohort was 7.4 months (range 1.2–98.9 months). At time of analysis, 101 of the 112 patients had died of disease.

The majority (83%) of patients (n = 93) were treated using a multiple field technique, 18 patients (16%) were treated with parallel-opposed lateral beams and one patient received a wedge pair technique. Radiotherapy was primarily delivered focally to the radiologically evident gross tumor volume with a 2–3 cm margin for the clinical target volume; however, three patients (3%) received a whole brain component because of multifocal disease. All but one of the 81 patients completed their hypofractionated course of radiation. This patient completed only 14 of 15 fractions due to intercurrent illness.

The majority (80%) of patients (n = 90) had a Karnofsky Performance Status (KPS) score of 70 or greater. The breakdown of KPS scores, as well as other patient characteristics can be seen in Table 1. Among the 57 concurrent RT + TMZ patients, the mean age was 70.2 years (range 60–86, SD 6.3), median KPS was 80 (range 30–100) and 24/57 (42%) had received prior debulking surgery in the form of either subtotal or gross total resection. The 55 patients treated without concurrent chemotherapy were similar in demographics with a mean age of 70.3 years (range 60–81, SD 5.6), median KPS of 70 (range 30–90), however the frequency of debulking surgery (subtotal or gross total resection) was greater, 34/55 patients (62%).
Table 1

Patient characteristics of elderly GBM patients

Characteristic

All patients

Concurrent chemotherapy

No concurrent chemotherapy

P-value

Salvage TMZ

RT alone

P-value

Number of patients

n = 112

n = 57

n = 55

 

n = 24

n = 31

 

Age (years)

   

0.943

  

0.025

 Mean

70

70

70

 

68

72

 

 Range

60–86

60–86

60–81

 

60–81

62–80

 

Age distribution

   

0.904

  

0.209

 60–64 years

24 (21%)

12 (21%)

12 (22%)

 

8 (33%)

4 (13%)

 

 65–69 years

29 (26%)

14 (25%)

15 (27%)

 

7 (29%)

8 (26%)

 

 70–74 years

30 (27%)

17 (29%)

13 (24%)

 

5 (21%)

8 (26%)

 

 ≥75 years

29 (26%)

14 (25%)

15 (27%)

 

4 (17%)

11 (35%)

 

Male sex

73 (65%)

37 (65%)

36 (65%)

0.952

16 (67%)

20 (65%)

0.868

Extent of surgery

   

0.045

  

0.378

 Biopsy

54 (48%)

33 (58%)

21 (38%)

 

7 (29%)

14 (45%)

 

 Subtotal resection

23 (21%)

7 (12%)

16 (29%)

 

7 (29%)

9 (29%)

 

 Gross total resection

35 (31%)

17 (30%)

18 (33%)

 

10 (42%)

8 (26%)

 

Steroid dose prior to RT (milligrams)

   

0.208

  

0.019

 Median

6

4

8

 

4

12

 

 Range

0–32

0–32

0–32

 

0–24

0–32

 

Karnofsky Performance Status (KPS)

       

 Median

70

80

70

 

80

70

 

 Range

30–100

30–100

30–90

 

30–90

30–90

 

KPS

   

0.063

  

0.001

 30: Severely disabled

3 (3%)

1 (2%)

2 (4%)

 

1 (4%)

1 (3%)

 

 40: Disabled

0 (0%)

0 (0%)

0 (0%)

 

0 (0%)

0 (0%)

 

 50: Considerable assistance

8 (7%)

5 (9%)

3 (5%)

 

0 (0%)

3 (10%)

 

 60: Occasional assistance

11 (10%)

7 (12%)

4 (7%)

 

2 (8%)

2 (6%)

 

 70: Unable to work

44 (39%)

14 (25%)

30 (55%)

 

7 (29%)

23 (74%)

 

 80: Normal with effort

33 (29%)

19 (33%)

14 (25%)

 

13 (54%)

1 (3%)

 

 90: Minor symptoms

12 (11%)

10 (18%)

2 (4%)

 

1 (4%)

1 (3%)

 

 100: Normal

1 (1%)

1 (2%)

0 (0%)

 

0 (0%)

0 (0%)

 

The median duration of symptoms prior to diagnosis was 1 month (range <1–12 months). Most patients presented with a combination of symptoms including seizures (31%), focal neurological deficits (58%), cognitive changes (52%), headaches (43%), or vision changes (7%). The majority (78%) of patients received magnetic resonance imaging (MRI) as pre-operative imaging although 25 patients (22%) only had computed tomography (CT). Enhancement was most commonly seen on imaging in 96% of cases (n = 107) followed by necrosis (16%) and calcification (5%). There was no difference in tumour laterality with 56% of patients having right-sided tumours and 39% having left-sided tumours. Five patients (4%) had tumours crossing midline to involve both hemispheres.

Concurrent temozolomide

Two (4%) of the 57 patients who received RT with concurrent TMZ did not complete their concomitant course of chemotherapy due to intercurrent illness. Forty of these 57 patients received adjuvant chemotherapy with nine patients (23%) requiring a dose reduction due to toxicity (n = 8) or unrelated illness (n = 1). Of the 40 patients receiving adjuvant therapy, 11 patients (28%) completed the full six cycles of adjuvant TMZ. The other 29 patients did not complete the six cycles of adjuvant chemotherapy due to tumour progression (n = 26), toxicity (n = 1), illness not related (n = 1), and unknown (n = 1). Of the 11 patients who completed six cycles of adjuvant TMZ, five continued to receive further adjuvant chemotherapy. Two patients received an additional three cycles of adjuvant TMZ, two patients received six further cycles and one patient received an additional 15 cycles of adjuvant TMZ chemotherapy. At the time of tumor progression, 7/57 (12%) patients received salvage surgery and 15/57 (26%) received additional salvage chemotherapy including temozolomide (n = 6), lomustine (CCNU) (n = 7), temozolomide and CCNU (n = 1) or dasatinib as per Radiation Therapy Oncology Group (RTOG) 0627 study (n = 1).

No concurrent temozolomide

Among those 55 patients treated initially with radiation alone, 24 patients received salvage temozolomide given at the time of progression. The other 31 patients did not receive any further salvage chemotherapy or radiation therapy except for one patient who had salvage surgery. The median time from completion of radiation to institution of salvage chemotherapy among those receiving salvage chemotherapy was 4.6 months (range 0.7–16.3 months). Six of the 24 (25%) patients had repeat surgery prior to institution of salvage chemotherapy. Salvage chemotherapy was instituted for 10 patients with asymptomatic radiographic progression and in 14 patients with both increasing clinical symptoms and evidence of radiographic progression. Temozolomide was the first choice for salvage chemotherapy in 22 of 24 cases. Two patients (treated early in the series) received lomustine (CCNU) for one cycle and two cycles, respectively prior to institution of salvage temozolomide. Salvage temozolomide was prescribed using the same schedule as the Stupp adjuvant TMZ regimen and a median of three courses were delivered (range 1–11 cycles). Other salvage chemotherapy regimens used in addition to temozolomide among the 24 patient salvage cohort included CCNU (n = 6), etoposide (n = 3), dasatinib as per RTOG 0627 study (n = 2) and a combination of CCNU plus tamoxifen (n = 1).

Progression-free and overall survival analysis

Median overall survival (OS) was 6.9 months (95% CI, 4.5–8.6) for the concurrent RT + TMZ cohort compared to 9.3 months (95% CI 5.9–11.8; Fig. 1) for the cohort without concurrent TMZ (P = 0.351) (Table 2). Sub-group analysis of the latter cohort revealed patients treated with salvage TMZ had increased median OS of 13.3 months (95% CI, 9.9–19.3) when compared to those without further treatment (median OS 5.7 months, 95% CI 3.8–8.9, (P = 0.012); Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs11060-011-0766-3/MediaObjects/11060_2011_766_Fig1_HTML.gif
Fig. 1

Progression-free (PFS) and overall survival (OS) Kaplan–Meier curves

Table 2

Comparison of progression-free (PFS) and overall survival (OS)

Characteristic

RT + concurrent TMZ

RT + no concurrent TMZ

P-value

RT + salvage TMZ

RT alone

P value

Number of patients

n = 57

n = 55

 

n = 24

n = 31

 

Median PFS—months (95% CI)

3.9 (2.9–5.3)

4.7 (3.2–6.1)

0.714

5.1 (3.2–6.5)

3.8 (2.9–8.0)

0.336

Median OS—months (95% CI)

6.9 (4.5–8.6)

9.3 (5.9–11.8)

0.351

13.3 (9.9–19.3)

5.7 (3.8–8.9)

0.012

https://static-content.springer.com/image/art%3A10.1007%2Fs11060-011-0766-3/MediaObjects/11060_2011_766_Fig2_HTML.gif
Fig. 2

Overall survival between hypofractionated radiation alone vs. hypofractionated radiation with salvage TMZ chemotherapy

Prognostic factors

Univariable analysis based on the overall survival experience limited to the first 12 months where the cohort survival curves are separated and Cox proportional hazards regression methodology is applicable, reveals concurrent chemotherapy (P = 0.003) and surgery less than a gross total resection (i.e. biopsy or STR) (P < 0.001) were unfavourable prognostic indicators. Multivariable analysis continues to show concurrent chemotherapy (P = 0.004) and limited extent of surgery (P = 0.001) remaining as independent unfavourable prognostic indicators of overall survival (Table 3).
Table 3

Uni- and multi-variable analysis of prognostic parameters for overall survival restricted to first 12 months

Parameter

Hazard ratio (95% CI)

P value

Univariate

  

 Agea

1.50 (1.03–2.18)

0.033

 Treatment modality

 

0.003

  Concurrent chemotherapy

2.73 (1.40–5.33)

 

  RT alone

3.37 (1.63–6.95)

 

  Salvage chemotherapy

Reference

 

 Karnofsky Performance Status score >70

0.72 (0.45–1.13)

0.154

 Extent of surgery

 

<0.001

  Biopsy

3.02 (1.69–5.40)

 

  Subtotal resection

2.47 (1.26–4.86)

 

  Gross total resection

Reference

 

Multivariate

  

 Treatment modality

 

0.004

  Concurrent chemotherapy

2.65 (1.34–5.24)

 

  RT alone

3.31 (1.60–6.86)

 

  Salvage chemotherapy

Reference

 

 Extent of surgery

 

0.001

  Biopsy

2.85 (1.59–5.12)

 

  Subtotal resection

2.67 (1.35–5.28)

 

  Gross total resection

Reference

 

aHazard ratio for age is in terms of decades

Treatment-related toxicities

Toxic effects were graded accordingly to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. A grading (severity) scale is provided for each adverse event (AE) term, ranging from Grade 1–4 corresponding to mild, moderate, severe, life-threatening or disabling AE. There were no Grade 5 adverse events in this patient series, that is, no death related to AE.

Sixteen of the 81 patients (20%) who received chemotherapy, either concurrent or as salvage, experienced some form of hematological toxicity, primarily with neutropenia and/or thrombocytopenia. Only seven of these patients had Grade 3 or 4 hematological toxicity; 5 patients (9%) among those receiving concurrent TMZ and two patients (8%) among those receiving salvage chemotherapy. Grade 3 or 4 toxicities included neutropenia (n = 3), thrombocytopenia (n = 2), or both (n = 2). One patient developed pneumonia during concurrent TMZ chemotherapy despite antibiotic prophylaxis.

Discussion

Treatment options for elderly GBM patients

The benefit of radiation therapy in elderly GBM patients has been proven in a phase III randomized control trial (RCT) by Keime-Guibert et al. Radiation therapy results in a modest but statistically significant improvement in survival in elderly patients with GBM compared to best supportive care alone, without reducing the quality of life or cognition [11].

The use of hypofractionated radiation has been previously studied in elderly patients in an attempt to reduce patient effort and inconvenience as well as improve quality of life. One of the earliest and largest series reported was the Johns Hopkins experience with the “SHORT regimen” using large or whole brain fields to treat to 30 Gy in 10 daily fractions. Following a two-week break, an additional 3.0 Gy × 7 fractions was delivered to a reduced cone-down field for a total dosage of 51 Gy given in 17 fractions over 5.5 weeks [12]. Kleinberg et al. concluded that this shortened regimen was an appropriate treatment option for most malignant glioma patients (RTOG groups IV–VI), resulting in similar median overall survival as standard regimens, ranging from 5 to 13 months, with acceptable acute side effects and low risk of brain necrosis. Most of the additional hypofractionated RT studies have used daily fractionation schemes ranging from 2.5 to 3.33 Gy per fraction to total dosages of 37.5 to 51 Gy; and reported similar survival rates ranging from 6.0 to 9.5 months [1317]. Hulshof et al. reported on their patient series treated with extreme hypofractionation schemes of 8 fractions × 5 Gy and 4 fractions × 7 Gy conformal irradiation, typically given to poor prognostic glioblastoma patients at the University of Amsterdam. Compared with conventional fractionation of 33 fractions × 2 Gy where median survival was 7 months, the hypofractionation groups provided equal palliation with comparable survival rates of 5.6 to 6.6 months [18].

Roa et al. [9] prospectively compared conventional radiotherapy (60 Gy in 30 fractions over 6 weeks) to an abbreviated course of RT (40 Gy in 15 fractions over 3 weeks) with overall survival as the primary endpoint, and found no statistical difference in survival. Similarly, Quong et al. [19] noted no survival difference in a randomized comparison of 35 Gy in 10 fractions compared to standard 60 Gy in 30 fractions.

The use of the Stupp regimen in elderly patients with glioblastoma has also been reported with median survivals in the range of 10.2–15.3 months, although no randomized comparisons against radiation alone are available (see Table 4) [2025]. In the original Stupp protocol, patients between the ages of 50–70 (RPA class V) were included and median survival among this population was 10 months [8].
Table 4

Selected series of elderly (≥60 years) glioblastoma multiforme patients treated with radiation and/or temozolomide[9, 11, 2038]

Author

Year

Type of study

Min age

Min KPS

N (total)

Med age

Med KPS

Treatment

Radiation dose and fractionation

N (by cohort)

Median survival (months)

Ewelt [32]

2010

Retrospective review

65

n/s

103

70.8

70

Surgery alone

n/a

31

2.2

Surgery + RT

60Gy/30#

37

4.4

RT + aTMZ

60Gy/30#

35

15.0*

Fiorica [20]

2010

Prospective study

65

n/s

42

71.3

70

RT + cTMZ

60Gy/30#

33

10.2**

RT + cTMZ

45Gy/15#

9

 

Gerstein [21]

2010

Retrospective review

65

n/s

51

70

n/s

RT + cTMZ

60Gy/30#

51

11.5

Kimple [22]

2010

Retrospective review

70

n/s

31

76

60

BSC

n/a

13

2.1

RT alone

60Gy/30#

4

7.1

RT + cTMZ

60Gy/30#

14

12.6*

Laigle-Donadey [29]

2010

Retrospective RCT analysis

70

n/s

39

75

n/s

TMZ alone

n/a 

39

9.0

Malmstrom [31]

2010

Phase III RCT

60

n/s

342

70

n/s

RT alone

60Gy/30#

n/s

6.0

RT alone

34Gy/10#

n/s

7.5

TMZ alone

n/a

n/s

8.0

Minniti [23]

2010

Prospective study 

70

70

83

73.2

80

RT + cTMZ (mMGMT)

60Gy/30#

42

15.3*

RT + cTMZ (umMGMT)

60Gy/30#

41

10.2

Muni [33]

2010

Prospective study

70 (50 if KPS≤70)

n/s

45

70

n/s 

RT alone

30Gy/6#

23

7.3

RT + aTMZ

30Gy/6#

22

9.4*

Wick [30]

2010

Phase III RCT

65

60

373

n/s

n/s

TMZ alone

n/a

n/s

n/s

RT alone

54–60 Gy

n/s

n/s

Brandes [34]

2009

Prospective study

65

n/s

58

68

80

RT + cTMZ

60Gy/30#

58

13.7

Minniti [35]

2009

Prospective study

70

60

43

73

70

RT + aTMZ

30Gy/6#

43

9.3

Combs [24]

2008

Retrospective review

65

n/s

43

67

n/s 

RT + cTMZ

60Gy/30#

43

11.0

Minniti [25]

2008

Prospective study

70

70

32

73.6

80

RT + cTMZ

60Gy/30#

32

10.6

Sijben [36]

2008

Retrospective review

65

n/s

39

70

67

70

80

RT alone

45Gy/15#

20

5.2

RT + cTMZ

45Gy/15#

19

8.5

Chamberlain [28]

2007

Prospective study

70

50

15

79

70

TMZ alone

n/a 

15

6.0

Keime-Guibert [11]

2007

Phase III RCT

70

n/s

85

73

73

75

BSC

n/a

42

4.2*

RT + BSC

50Gy/28#

39

7.3

Chinot [27]

2004

Phase II study

70

60

32

75

70

TMZ alone

n/a

32

6.4

Roa [9]

2004

Phase III RCT

60

n/s

100

72.4

71

70

70

RT alone

60Gy/30#

 

5.1

RT alone

40Gy/15#

5.6

Brandes [37]

2003

Prospective study

65

60

79

70

72.5

RT alone

59.44Gy/33#

24

11.2

69

80.0

RT + aPCV

59.44Gy/33#

32

12.7

68

77.0

RT + aTMZ

59.44Gy/33#

22

14.9*

Glantz [26]

2003

Retrospective review

70

n/s

86

73.3

67.4

RT

60Gy/33#

54

4.1

74.5

67.7

TMZ alone

n/a 

32

6.0

Bauman [38]

1994

Prospective study

65

50

29

69.2

52.4

RT alone

30Gy/10#

29

6

Min minimum, KPS Karnofsky Performance Status, Med median, RCT randomized control trial, RT radiotherapy, aTMZ adjuvant temozolomide chemotherapy, cTMZ concurrent temozolomide chemotherapy, mMGMT methylated O-6-methylguanine-DNA methyltransferase, umMGMT unmethylated O-6-methylguanine-DNA methyltransferase, aPCV adjuvant procarbazine, lomustine (CCNU), and vincristine chemotherapy, n/s not stated, n/a not applicable

* Statistically significant (P < 0.05), ** Total median survival (months)

An alternative treatment paradigm to radiotherapy or combined treatment involves the use of chemotherapy alone in elderly GBM patients and has been proposed by those concerned with potential radiation-induced neurological and cognitive side-effects. Glantz et al. [26] first reported a retrospective study of 86 consecutive elderly (age ≥70) patients with similar survivals when treated with either standard-course RT (60 Gy in 33 fractions) or monthly temozolomide alone (median survival, 4.1 vs. 6.0 months, P = 0.198). Chinot et al. [27] subsequently reported a phase II study of temozolomide (adjuvant Stupp regimen) without radiotherapy in elderly (age ≥70) GBM populations with only mild adverse events, no observed neurotoxicity and comparable survival (median OS 6.4 months). Chamberlain et al. [28] also revealed similar survival rates (median 6 months) in a pilot study of primary temozolomide with a modified dosing schedule (42 days on; 14 days off). Laigle-Donadey et al. retrospectively analyzed patients who were eligible for the Keime-Guibert phase III RCT, but who refused to participate and ultimately treated with TMZ alone. Thirty-nine GBM patients (ages ranging 70–83 and median KPS 70) were treated up-front with oral TMZ with a mean of 5 cycles and a median overall survival of 9 months [29]. Taken together these studies suggest that primary chemotherapy may be a reasonable alternative to primary radiotherapy for this patient population.

More recently, two prospective Phase III randomized controlled trials comparing temozolomide alone in elderly GBM patients with standard-course RT and hypofractionated RT have been reported. The NOA-08 trial of the Neurooncology Working Group (NOA) of the German Cancer Society compared standard postsurgical involved-field RT to a total dosage of 54 to 60 Gy in malignant glioma patients greater than 65 years with minimum KPS 60 to dose-intensified temozolomide alone (one week on/one week off) [30]. The trial failed to show the non-inferiority of TMZ alone compared to RT alone in the primary treatment of elderly patients with malignant glioma and the rate of adverse and serious adverse events were higher in the TMZ arm. Of note, this trial was designed as a cross-over design to test the sequencing of these two treatments, however, few patients were well enough at first progression for crossover to the alternate therapy. Malmstrom et al. [31] reported a European randomized trial comparing standard-course RT (60 Gy in daily 2 Gy fractions over 6 weeks), hypofractionated RT (34 Gy in daily 3.4 Gy fractions over 2 weeks), and monthly TMZ alone. The results of this trial demonstrated no significant difference in overall survival between the three treatment arms, with median survivals of 6, 7.5, and 8 months, respectively (P = 0.14). No details regarding toxicity or the use of salvage therapies among the three treatment arms were provided.

Taken in this context, our population of patients treated with concurrent and adjuvant temozolomide experienced PFS and OS consistent with other series of radiation or chemotherapy alone. Additionally, among our patients treated with radiation alone, those receiving salvage temozolomide appeared to derive a significant benefit with a median overall survival greater than a year. Thus, based on our series, it would appear that for older patients with glioblastoma, a sequential strategy may allow the selection of patients who are more likely to benefit from combined modality treatment.

One potential explanation for the apparent lack of efficacy of concurrent temozolomide may be the shortened duration of overlap between the concurrent chemotherapy and radiotherapy. Hypofractionated radiotherapy (15 fractions) results in a 50% reduction in administered concurrent temozolomide (usually 42 consecutive days) that may have not been enough duration to provide meaningful radiosensitization or interaction and subsequent translation into clinical benefit. Alternatively, the benefit of temozolomide chemotherapy may be derived from the adjuvant setting rather than the concurrent, as this distinction in benefit remains undetermined even in the patient population eligible for the Stupp regimen (Table 4). Selected series of elderly (≥60 years) GBM patients treated with radiation and/or temozolomide [9, 11, 2038].

Toxicity of treatment

In our retrospective series, there were seven cases of Grade 3 or 4 hematological toxicity related to chemotherapy; five cases among those receiving concurrent and/or adjuvant chemotherapy and two patients in the salvage chemotherapy cohort. Others have reported significant Grade 3 or 4 hematologic toxicity in elderly GBM patients treated with temozolomide ranging from 21 and up to 42% [21, 29, 33, 36]. Our overall incidence of 8.64% Grade 3–4 hematological toxicity is lower than these other series. Our data thus suggests that the combined regimen has very acceptable toxicity, but casts much doubt on its efficacy. There were no reported cases of radiation necrosis.

Study limitations

This review has the usual limitations expected of a retrospective study. While the patient populations were similar in age and performance status at treatment, there was an imbalance in the prevalence of debulking surgery in the sequential therapy group. We were limited in our ability to abstract more subtle prognostic variables (such as co-morbidities, functional reserve, and overall fitness) and account for their influence on treatment assignment and outcome. For instance, sequentially treated patients would have been assumed to have responded sufficiently well to initial therapy to be eligible for salvage therapy. On the other hand, patients selected for immediate concurrent chemotherapy presumably were judged to be “fit” for combined therapy.

A graphical representation of the distribution of treatment strategy over the past 10 years at our institution is shown in Fig. 3. It reveals that hypofractionated radiation therapy alone was used in the first part of the decade and shows the use of concurrent temozolomide coincides roughly with the publication of the Stupp trial results. Subsequently, the use of concurrent temozolomide with hypofractionated radiation became equivocal with the interim analysis of our database and the opening of the NCIC CE.6/EORTC 26062-22061 trial, a randomized phase III study comparing TMZ and short-course RT versus short-course RT alone for elderly GBM patients who are assessed as not suitable for the Stupp regimen. The clustering of patients by year according to treatment suggests it is less likely that there are subtle selection biases, but may rather represent a temporal bias where our group made an institutional decision to change treatment and management.
https://static-content.springer.com/image/art%3A10.1007%2Fs11060-011-0766-3/MediaObjects/11060_2011_766_Fig3_HTML.gif
Fig. 3

Timeline of patient cases based on date of diagnosis

Conclusion

Our results demonstrate that the strategy of prescribing immediate concurrent temozolomide chemotherapy with hypofractionated radiation does not appear to confer a significant survival benefit over radiotherapy alone in the elderly patient population studied. A sequential approach using hypofractionated radiotherapy initially with temozolomide reserved for salvage at time of progression may be a more effective strategy allowing for better selection of patients with sensitive disease who may benefit from combined therapy and may be better tolerated. The NCIC CE.6/EORTC 26062-22061/TROG 08-02 study will help answer this question in a definitive fashion. This trial is a randomized Phase III study of temozolomide and short-course radiation versus short-course radiation alone in the treatment of newly diagnosed GBM in elderly patients (minimum age 65 years). Outside of the trial, we utilize a sequential hypofractionated radiotherapy and chemotherapy strategy for those patients who are felt not to be fit for the standard Stupp regimen. For older “fit” patients, a standard Stupp regimen or randomized trials of conventional radiotherapy and temozolomide with other investigational agents are offered. Future studies that would assist decision making in this patient population would include correlation of treatment outcomes with standardized assessment of “fitness” among older glioblastoma populations using tools such as comprehensive geriatric assessment scales [39].

Acknowledgments

The authors would like to thank Ms. Frances Whiston and Mr. Larry Stitt, M.Sc., from the Clinical Research Unit, London Regional Cancer Program, for assistance with statistical analysis.

Conflict of interest

The authors declare no potential conflicts of interest relevant to this article.

Copyright information

© Springer Science+Business Media, LLC. 2011