- Email: [email protected]
Repeat-surgery at Glioblastoma recurrence, when and why to operate?
Clinical Neurology and Neurosurgery 136 (2015) 89–94
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Repeat-surgery at Glioblastoma recurrence, when and why to operate? Genevieve Ening a,b,∗ , Mai Thi Huynh b , Kirsten Schmieder a,b , Christopher Brenke a,b a Department of Neurosurgery, Knappschafts-Krankenhaus Bochum-Langendreer, Ruhr-University of Bochum, In der Schornau 23-25, 44892 Bochum, Germany b Department of Neurosurgery, University Medical Center Mannheim, University of Heidelberg, Theodor-Kutzer-Ufer 1, 68167 Mannheim, Germany
a r t i c l e
i n f o
Article history: Received 14 April 2015 Received in revised form 20 May 2015 Accepted 21 May 2015 Available online 28 May 2015 Keywords: Glioblastoma Survival Recurrence Repeat surgery Complications
a b s t r a c t Objective: Glioblastoma (GB) recurrence is inevitable; guidelines for treatment at disease recurrence are deficient. Clinicians are faced with deciding whom to choose for repeat-surgery. This study analyzes recurrence therapy modalities, investigates characteristics of patients operated on at recurrence and evaluates outcome benefit. Methods: Consecutive adult patients operated on for de novo GB at a single institution from 2006 to 2011 were reviewed. Clinical, radiographic and molecular data of 141 patients diagnosed of recurrent disease were assessed. Reasons for recurrence therapy and therapy modalities were reviewed. Univariate analysis was used to analyze differences in parameters of patients operated on at recurrence and those not. Impact of re-operation on survival was evaluated by the Kaplan-Meier method and Log-rank test. Results: 53 (38%) patients were selected for repeat surgery upon recurrent disease, this was followed by either chemotherapy (CT) (40%), radiotherapy (8%) or both (49%). 57 (40%) patients received CT alone, which was the most frequent mono-second-line therapy opted for. Most frequent indications for repeatsurgery were maximum possible tumor resection mass reduction and symptom relief (62% and 21%, respectively). Univariate analysis of re-operated vs. not operated patients, showed significant differences for age (p = 0.0001* ) and Karnofsky Performance status (KPS) >70 at both primary and repeat tumor resection (p = 0.013* and 0.0001* , respectively). The operated group had a significantly lower Charlsoncomorbidity-index ≤ 3 (p = 0.004* ) and larger tumor size (p = 0.0001* ). Complication risk at recurrence was not significantly different between groups (p = 0.069). However, patients chosen for repeat surgery had significantly less complications at index surgery (p = 0.006* ). Median time from recurrence to death was 11 months (range, 1–33 months) for operated patients as opposed to 5 months (range, 0–22 months) for not operated patients. The former survived significantly longer; 19 months compared to 13 months for those not operated upon (p = 0.002* ). Conclusions: Our study depicts that patients eligible for repeat-surgery at GB recurrence are characterized by a KPS > 70% before primary and repeat-surgery, Charlson-comorbidity-index ≤ 3, large tumor size and young age. These well-selected patients survive significantly longer after repeat-surgery without being at a higher complication risk in comparison to patients not operated upon. © 2015 Published by Elsevier B.V.
1. Introduction Recurrent disease in Glioblastoma (GB) patients is almost ubiquitous irrespective of improved therapy for newly diagnosed GB [1]. Over the past decade, treatment strategies for recurrent patients
∗ Corresponding author at: Department of Neurosurgery, Ruhr-University Bochum, In der Schornau 23-25, D-44892 Bochum, Germany. Tel.: +49 23429980255; fax: +49 2342993609. E-mail addresses: [email protected], [email protected] (G. Ening). http://dx.doi.org/10.1016/j.clineuro.2015.05.024 0303-8467/© 2015 Published by Elsevier B.V.
have become more aggressive, patients are offered various options including repeat-surgery, re-radiation, salvage chemotherapy [2] and alternating electric fields, so called tumor-treating fields [3]. Although some authors report of improved survival after repeatsurgery of recurrent tumor [4], the indications for, and efficacy of repeated surgery at time of tumor recurrence are not well defined [5]. In general, evidence based guidelines for treatment decision-upon disease recurrence are lacking [1]. Deciding to opt for repeated surgery at recurrence poses a dilemma for physicians, patients and their families. We reviewed consecutive patients with GB and sought to answer following questions:
90
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
What therapy modalities were decided on at recurrence? When was second-line therapy performed? What were the reasons for or against repeat-surgery? Which patient, tumor and first-line treatment characteristics differentiated the patients re-operated on from those not operated on at tumor recurrence? How does repeat-surgery impact patient outcome?
2. Methods 2.1. Patient selection Between 2006 and 2011 patients >18 years diagnosed of GB at a single academic institution (university hospital Mannheim, Germany) were identified by retrospective review of the neurosurgical department’s database. Institutional review board approval was obtained for all aspects of the study and all patients consented to their medical records being accessed for the study. Tumor diagnosis was determined by a senior neuropathologist according to the WHO classification of central nervous system tumors [6]. Molecular analysis was performed in the same neuropathological department. Tumor recurrence was defined as reappearance in the contrast enhanced magnetic resonance image (MRI) according to the Revised Assessment in Neuro-Oncology (RANO) criteria [7].
2.2. Recorded variables The type of second-line therapy employed; re-surgery, chemotherapy (CT) radiotherapy (RT) or combination of the various modalities was recorded. Furthermore, the times at which therapy was begun and reasons for deciding for or against repeat surgery were reviewed. Demographic parameters included age and gender. Patient variables comprised preoperative functional status, indexed by the Karnofsky Performance status (KPS) and dichotomized as >70% or ≤70%. Preoperative comorbidity at primary tumor surgery was classified by the Charlson-comorbidity index [8] as described elsewhere [8,9] and was further dichotomized as >3 or ≤3. Contrast enhanced MRIs were obtained preoperatively, <48 h after tumor operation (at both primary and repeat surgery), and during follow-up. Two independent reviewers blinded to patient outcomes carried out assessments. Tumor characteristics documented included anatomical tumor location as described previously [10]. In summary, multifocal lesions were defined as multiple separate enhancing lesions with or without connection. A deep lesion was defined as a lesion located in the basal ganglia, the thalamus, the corpus callosum, and/or the brainstem. Eloquent brain was stated as the motor, sensory, speech, or visual cortex, the basal ganglia/internal capsule, the thalamus, or the brainstem. Volumetric analysis including the contrast-enhancing preoperative tumor volume (CE-PTV) and the extent of resection (EOR) were measured as described elsewhere [11], accordingly gross total resection (GTR), subtotal resection (STR) and partial resection (PR) was defined as 100%, >95% <100% or <95% tumor resection, respectively. Molecular analysis was carried out according to standard methods for DNA extraction from formalin-fixed paraffin-embedded tumor tissue or from unfixed frozen tumor tissue samples. O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation analysis was carried out by methylation-specific polymerase chain reaction. Isocitrate dehydrogenase (IDH)-1 codon 132 was analyzed by direct sequencing. In cases of ambiguous results, the IDH1 sequences were amplified by a different set of primers [12,13].
Fig. 1. Bar chart illustrating the different therapy modalities employed at recurrence and the frequency of implementation.
All recorded complications were categorized into neurological (N), regional/surgical (S) and systemic/medical (M) complications, as suggested for patients undergoing craniotomy by the Glioma Outcome Project [14]. Patients were followed up clinically and by MRI every 3 months. Vital status (alive or dead) and time of death was obtained through patient records and information from relatives or treating physicians. Death data was updated through to June 2012. Primary study endpoints were overall survival (OS) from the date of surgery of primary tumor and time to progress after repeat surgery or therapy at recurrence. In cases where death could not be confirmed by any means, the patients were classified as lost to follow-up at the time of their last clinic visit. 2.3. Statistical analysis Statistical analyses were performed with SPSS version 22.0 (SPSS Inc., Chicago, Illinois, USA). Descriptive data was presented as means ± SDs for parametric data and medians with ranges for nonparametric data. Intergroup frequencies were compared using the Fisher exact test/Mann-Whitney U-test for categorical variables and the Student t-test for continuous variables. Two-sided p values less than 0.05 were considered statistically significant. Factors independently associated with choice for re-operation were assessed by univariate analysis. The Student t-test was used for continuous variables and the chi-square for categorical variables. The Kaplan–Meier-method and Log-rank analysis was used to assess the impact of repeat-surgery on outcome. 3. Results Of the all patients reviewed, 18 died within 30 days after surgery of the initial tumor. 43 died without having been diagnosed of a recurrent tumor. 31 patients had no reliable data and were removed from the cohort. 141 patients were diagnosed of a recurrent tumor and made up our final study group. Fig. 1 illustrates the various types of therapy modalities employed for the 141 cases. 57 patients (40%) received CT alone, this was the most frequent mono-second-line therapy decided on. 53 (38%) of the patients were re-operated on at tumor recurrence. Histological analysis showed that whilst 43 patients had vital recurrent tumor, 10 patients showed no vital tumor. Hence, were viewed to have pseudoprogression. Also, out of the 53 operated cases, 49% received RT and CT, (32%) CT and (8%) RT alone following repeat-surgery. Only six patients (4%) did not receive further therapy following repeat surgery. This was due to patient refusal in five cases and deteriorated condition following a surgical complication after repeat surgery in one case. Fig. 2 shows the time point at which second-line therapy was implemented.
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
91
Table 1 Summary of clinical, treatment, tumor and imaging characteristics of 141 patients with recurrent glioblastoma and univariate analysis of factors in relation to surgery of recurrent tumor. Parameter
Patients No. (%) Demographics Age (mean +-SD)
Fig. 2. Bar chart depicting the time points of recurrence treatment commence and corresponding number of patients.
In 70 cases, more than 50% of patients’ recurrent tumor was distinctly diagnosed via radiographic assessment and second-line therapy was commenced immediately hereafter. In contrast hereto for 3 cases recurrence was unclear from radiographic imaging therefore time elapsed before second-line therapy initiation. No exact specification for the time point of second-line therapy commence was extractable in 15 cases. In 18 patients CT/RT was performed as first option, upon further radiographic progress reoperation was then decided upon. In 22 cases, repeat-surgery was decided on directly at diagnosis of recurrence. Fig. 3 sums up the reasons for deciding to operate patients at recurrence. In most cases (33) the aim of re-operation was firstly to attain maximum possible tumor resection and reduce tumor mass and secondly to relieve symptoms (11 cases). In 3 cases, initial decision of CT/RT was changed to repeat-surgery as the former therapy modalities were not tolerated by the patients. 38 patients were in a bad clinical condition for surgery and of these best supportive care was offered in 9 cases. As illustrated in Fig. 1 8 patients had recurrent tumor locations not eligible for surgery. The mean time interval between operations was 9 ± 6.5 months. 3.1. Univariate analysis of characteristics Table 1 summarizes demographic, patient, tumor and clinical parameters of the 141 patients with recurrent tumors. 53 (38%) patients underwent surgery and 88 (62%) were treated otherwise Fig. 1. The two groups had similar gender distribution (p = 0.15). Patients operated on were significantly younger mean 55.4 years compared to 63.7 years (p = 0.0001* ). Also, patient characteristics (KPS ≥ 70, CCI < = 3) at index surgery were significantly distinct between groups (p = 0.013* and 0.004* , respectively). This was also the case for KPS ≥ 70, at recurrent tumor operation (p = 0.0001* ). Tumor and molecular characteristics (anatomic and eloquent location, CE-PTV, EOR, MGMT and IDH 1 status at index surgery proved no significant difference between the two groups. Although there was no difference in initial tumor volume (CE-PTV) in the operated and not operated groups, patients who were operated on at recurrence had significantly larger recurrent tumor volumes 28.6 cm3 vs. 15.4 cm3 (p = 0.006* ). Group comparison with regard to first-line therapy; combined radiochemotherapy according to the EORTC-Protocol or other therapy modalities showed no significant difference (p = 0.062). Patients operated on at recurrence had significantly less complications at index surgery compared to those not operated (p = 0.006* ). However, there was no significant difference in recurrence treatment compliation rate between those chosen for repeat surgery and those who were not (p = 0.069).
No repeatsurgery
Repeatsurgery
88 (62)
53 (38)
63.7
55.4
Univariate analysis p Value
0.0001* 0.51
Sex Male Female
47 (62) 41 (63)
29 (38) 24 (37)
Patient characteristics KPS at index surgery <70 >70
18 (86) 70 (58)
3 (14) 50 (42)
KPS at tumor recurrence <70 >70
44 (86) 34 (46)
7 (14) 40 (54)
CCI at index surgery ≤3 >3
42 (53) 46 (75)
38 (47) 15 (25)
0.013*
0.0001*
0.004*
Tumor Characteristics Anatomical tumor localization at index surgery 25 (57) Frontal Temporal 21 (60) 11 (61) Parietal Occipital 2 (40) 26 (72) Not localized to one site 3 (100) Others
19 (43) 14 (40) 7 (39) 3 (60) 10 (28) 0 (0)
Eloquent tumor localization at index surgery 12 (67) Eloquent 50 (70) Near-Eloquent Non-Eloquent 26 (51)
6 (33) 22 (31) 25 (49)
0.411
0.106
Tumor volume CE-PTV after first operation CE-PTV (cm3 ) 33.13
36.48
0.573
Tumor volume CE-PTV at tumor recurrence 15.37 CE-PTV (cm3 )
28.57
0.016*
EOR initial surgery <95% >95% < 100% 100%
22 (56) 20 (54) 22 (69)
17 (44) 17 (46) 10 (31)
MGMT Methylated Unmethylated
26 (54) 36 (61)
22 (46) 23 (39)
IDH1 Mutant Wildetype
2 (33) 66 (64)
4 (67) 38 (36)
Clinical characteristics Complication after index surgery Yes No
60 (52) 28 (49)
24 (32) 29 (51)
Complication after recurrent operation 23 Yes 65 No
21 32
First-line adjuvant therapy RT/TMZ → TMZ Other
0.419
0.302
0.149
0.006*
0.069
0.062 66 (59) 22 (86)
46 (41) 7 (14)
Abbreviations: CE-PTV-contrast-enhancing preoperative tumor volume, MGMTO6-methylguanine-DNA methyltransferase, IDH-1-Isocitrate dehydrogenase, EORextent of resection, GTR-gross total resection, STR-subtotal resection, PR partial resection PR, KPS Karnofsky Performance status and CCI Charlson-comorbidityindex, RT/TMZ → TMZ Concomitant Radiochemotherapy with Temozolomide (TMZ) and following TMZ therapy.
92
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
Fig. 3. Bar chart showing the frequencies of reasons for deciding to opt for repeat-surgery in patients.
3.2. Outcome The median time interval from diagnosis of recurrent disease to death was 11 months (range, 1–33 months) for operated patients as opposed to 5 months (range, 0–22 months) for not operated patients. Patients operated on at recurrence survived significantly longer 19 months compared to 13 months for those not operated upon (p = 0.002* ). This is illustrated by the Kaplan-Meier curve in Fig. 4. Evaluation of first-line therapy employed in patients with histological evidence of vital tumor as compared to those diagnosed of pseudoprogression showed that 37 out of the 43 patients with vital recurrent tumor received initial combined radiochemotherapy. In contrast, 9 out of the 10 patients were classified to have a pseudoprogression. Kaplan-Meier outcome analysis showed no difference in outcome with regard to patients with vital recurrent tumor and those with pseudoprogression. 4. Discussion 4.1. Therapy modalities decided on at recurrence. Whereas 91% of patients received some sort of therapy at recurrence, 9% of patients received no tumor specific treatment. In our study, 40% of patients received CT alone. In contrast to these, 38% of patients were chosen for repeat surgery. Of the latter, only 2 patients received surgery as a mono-therapy. This is in accordance with literature, where it is stated that for most patients, CT remains the treatment of choice at recurrence and for patients with a substantially reduced performance status, supportive care measures alone may be more appropriate [1]. Similar to treatment at primary tumor diagnosis, at recurrence, re-surgery was a therapy opted for in combination with other therapy modalities. Previous reports state that surgery as a mono-therapy in GB treatment is not reasonable and has to be combined with other therapies [16]. Hence, repeat-surgery seems a just therapy if further therapy options are available. 4.2. Time point for second-line therapy In 70 cases from our study, recurrent diagnosis was well-defined according to MRI imaging criterion and hence second-line therapy was commenced without delay. Three patients were observed initially and decided on treatment after further MRI scanning. In 22 patients, surgery was the initial second-line therapy option. In contrast, 18 patients received CT/RT at second-line and were
then re-operated upon response failure. This implies that, at tumor recurrence although the decision to offer second-line therapy is clear there is often uncertainty as to which therapy modality to opt for. Although some scores have been defined to aid decision making [16], they are not well established in clinical routine. Similar to most published studies in this field, our review also lacks data on quality of life [2]. In order to account for patients and relatives opinion regarding the time point to commence second-line therapy, this outcome measure may be of importance and should be considered in future studies. 4.3. Reasons for or against repeat-surgery Deciding to operate patients at recurrence was to attain maximum tumor resection allowing tumor mass effect reduction in 62% cases and to achieve symptom relief in 21% cases. Moreover, deteriorated patient condition was most often the reason why surgery was not opted for at recurrence (38 cases). In 10 cases, therapies other than repeat surgery were deemed more appropriate. We meticulously reviewed all protocols of the interdisciplinary neurooncological tumor board, this provided data on the reasons for or against repeat-surgery in each patient. In cases were emergency surgery was decided upon due to mass effect, the information was obtained from the emergency admission protocols. Furthermore, univariate analysis showed that irrespective of the tumor size at index surgery, at recurrence patients with significantly larger tumors were chosen for repeat-surgery. According to literature reports, mass reduction and symptom relief are the main aims for surgery at primary tumor diagnosis [17]. This implies that, reasons for patient repeat-surgery at recurrence are comparable to those for index surgery but still have an independent value. 4.4. Patient, tumor and treatment characteristics for repeat-surgery and impact on outcome Our results show that patients selected for repeat surgery are significantly younger, have a better functional status and less comorbidities. This may reflect some sort of selection bias favoring good candidates for repeat-surgery. To reduce this effect we dichotomized parameters such as KPS and CCI, lowering reviewer subjectivity. The confounding effect of age and performance status was investigated by Stark et al. Their data suggests that re-surgery should be considered in all patients with favorable performance status regardless of age [18]. Also Park et al. in trying to develop a preoperative scoring system, the NIH recurrent GBM scale, considered KPS but not age to be a more relevant parameter [16].
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
93
Fig. 4. Kaplan-Meier curve depicting significant difference in overall survival of the repeat-surgery group median of 19 months compared to not operated group 13 months (p = 0.002* ).
The results of our study clearly show that patients who are reoperated at recurrence survive longer after recurrence (11 months) and have a longer OS of 19 months. This implies an outcome benefit for repeat-surgery. Our findings of a 9 months mean time interval between operations accords with previous results where a time interval of at least 6 months between operations proved to be an important predictor of benefit from reoperation. {Hervey-Jumper, 2014 #3206}Moreover, patients chosen for repeat surgery had had a significantly less complication rate at index surgery (p = 0.006* ), this effect was no more evident at repeat-surgery (p = 0.069). This relates to previous results which state that for recurrent GB, an improvement in overall survival can be attained when the EOR is beyond 80%, emphasizing that this survival benefit must be balanced against the risk of neurological morbidity, which does increase with more aggressive cytoreduction. However, this is only in the early postoperative period [19]. This early postoperative neurological deterioration may be explained by findings from other studies where it was shown that rather than cortical or subcortical structural damage of eloquent brain tissue alone, peri- or postoperative ischemic lesions play a crucial role in the development of surgery-related motor deficits [20]. This suggests that the complication rate at recurrent surgery is irrespective of complications at primary surgery and each surgery has an individual complication risk. Hence, when deemed eligible patients should be chosen for repeat surgery despite the fact that they experienced complications at initial tumor resection. However, to keep complication rate at a minimum, modern strategies of maximal safe resection 5-aminolevulinic acid (ALA), intraoperative monitoring and intraoperative MRI should be employed. Additionally, in defining eloquent brain, areas directly adjacent to the M1 and/or M2 segments of the middle cerebral artery have not traditionally been considered as eloquent brain regions. Hence, injury to the above named vessels during surgery can result in damage to the eloquent brain regions they supply [21]. This difference
in anatomical and functional definition of eloquent regions may explain the insignificant correlations observed in our study as we made only an anatomical distinction of eloquent regions. Future clinical studies have to consider these different definitions of eloquence in order to better reflect the surgical situation. Our findings depict that patients with larger tumors at recurrence measured by volume of contrast enhancing lesion, were more likely to be chosen for repeat surgery and that smaller recurrent tumors were selected for other therapy modalities. This observation was not evident for tumor volumes at index surgery. In contrast, all other tumor characteristics, lobe location, eloquent location and EOR did not differ significantly between groups neither at initial tumor resection nor at recurrence Table 1. These results contradict findings of Bloch et al. and Quick et al. who showed that EOR at recurrence and not at initial surgery impacted survival, suggesting that gross total resection at second craniotomy could overcome the effect of an initial subtotal resection [22,23]. However our findings on CE-PTV somehow relate to reports which state the CE-PTV to be the more significant predictor of survival compared to EOR [24]. Also, it is reported that the resection cavity underestimates the volume of resected tissue and 5-ALA complete resections go significantly beyond the volume of pre-operative contrast-enhancing tumor bulk on MRI [25]. These contradicting volumetric analysis results may be a reflection of the persisting heterogeneity in volumetric measurements employed, and emphasizes the uncertainty and dilemma in defining the value of EOR for both initial and recurrent tumor resections [2]. Hence, future studies will have to consider various pre- and postoperative volumetric definitions in order to allow better comparison of result. 4.5. Study strengths and limitations Our data was obtained from a single institution making up a small cohort size, which in turn minors the statistical analysis.
94
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
However, this has the advantage of a single institutional data source reducing possible confounding effects of differences in the access to health care services between health centers. Although limited by the retrospective study design the pseudoprogression rate of 19% in our study cohort correlates with numerous previous reports [15] hence allowing representative analysis and conclusions. In differentiating between recurrence diagnosis and therapy induced pseudoprogression, modern imaging techniques, such as MRSpectroscopy or FET-PET analysis may have been of assistance. Unfortunately at the time of our study these techniques were up coming at our institution and not part of clinical routine. Hence we do not have comprehensive data on this account.” Furthermore, many retrospective studies published lack molecular data correlations [2] one strength of our study is the assessment of MGMT methylation status and IDH1 mutation status in correlation to repeat-surgery at recurrence. However, we did not see any significant correlation between these molecular parameters and decision for repeat-surgery. This may be explained by the fact that the strong predictive effect of MGMT methylation status on outcome is associated with response to CT [5] and not surgical treatment per se. 5. Conclusion In our study patients chosen for repeat-surgery at GB recurrence were of a younger age, showed a KPS > 70% before primary and repeat-surgery, had less comorbidity (CCI ≤ 3) and a larger tumor size compared to not operated patients. Patients selected by these criteria significantly survived longer after repeat-surgery without being at a higher risk for encountering complications. References [1] Preusser M, de Ribaupierre S, Wohrer A, et al. Current concepts and management of glioblastoma. Ann Neurol 2011;70:9–21. [2] Hervey-Jumper SL, Berger MS. Reoperation for recurrent high-grade glioma: a current perspective of the literature. Neurosurgery 2014;75:491–9. [3] Stupp R, Wong ET, Kanner AA, et al. NovoTTF-100A versus physician’s choice chemotherapy in recurrent glioblastoma: a randomised phase III trial of a novel treatment modality. Eur J Cancer 2012;48:2192–202. [4] Helseth R, Helseth E, Johannesen TB, et al. Overall survival, prognostic factors, and repeated surgery in a consecutive series of 516 patients with glioblastoma multiforme. Acta Neurol Scand 2010;122:159–67. [5] Stupp R, Brada M, van den Bent MJ, et al. High-grade glioma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2014;25(Suppl. 3), iii93-101. [6] Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 2007;114:97–109.
[7] van den Bent MJ, Wefel JS, Schiff D, et al. Response assessment in neurooncology (a report of the RANO group): assessment of outcome in trials of diffuse low-grade gliomas. Lancet Oncol 2011;12:583–93. [8] Balducci M, Fiorentino A, De Bonis P, et al. Impact of age and co-morbidities in patients with newly diagnosed glioblastoma: a pooled data analysis of three prospective mono-institutional phase II studies. Med Oncol 2012;29: 3478–83. [9] Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–83. [10] Lacroix M, Abi-Said D, Fourney DR, et al. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001;95:190–8. [11] Shi WM, Wildrick DM, Sawaya R. Volumetric measurement of brain tumors from MR imaging. J Neurooncol 1998;37:87–93. [12] Hartmann C, Hentschel B, Simon M, et al. Long-term survival in primary glioblastoma with versus without isocitrate dehydrogenase mutations. Clin Cancer Res 2013;19:5146–57. [13] Pusch S, Sahm F, Meyer J, et al. Glioma IDH1 mutation patterns off the beaten track. Neuropathol Appl Neurobiol 2011;37:428–30. [14] Chang SM, Parney IF, McDermott M, et al. Perioperative complications and neurological outcomes of first and second craniotomies among patients enrolled in the Glioma Outcome Project. J Neurosurg 2003;98:1175–81. [15] Brandes AA, Tosoni A, Spagnolli F, et al. Disease progression or pseudoprogression after concomitant radiochemotherapy treatment: pitfalls in neurooncology. Neuro Oncol 2008;10:361–7. [16] Park JK, Hodges T, Arko L, et al. Scale to predict survival after surgery for recurrent glioblastoma multiforme. J Clin Oncol 2010;28:3838–43. [17] Gulati S, Jakola AS, Nerland US, et al. The risk of getting worse: surgically acquired deficits, perioperative complications, and functional outcomes after primary resection of glioblastoma. World Neurosurg 2011;76:572–9. [18] Stark AM, Hedderich J, Held-Feindt J, et al. Glioblastoma—the consequences of advanced patient age on treatment and survival. Neurosurg Rev 2007;30:56–61, discussion 61-52. [19] Oppenlander ME, Wolf AB, Snyder LA, et al. An extent of resection threshold for recurrent glioblastoma and its risk for neurological morbidity. J Neurosurg 2014;120:846–53. [20] Gempt J, Krieg SM, Huttinger S, et al. Postoperative ischemic changes after glioma resection identified by diffusion-weighted magnetic resonance imaging and their association with intraoperative motor evoked potentials. J Neurosurg 2013;119:829–36. [21] Chang EF, Smith JS, Chang SM, et al. Preoperative prognostic classification system for hemispheric low-grade gliomas in adults. J Neurosurg 2008;109:817–24. [22] Bloch O, Han SJ, Cha S, et al. Impact of extent of resection for recurrent glioblastoma on overall survival: clinical article. J Neurosurg 2012;117: 1032–8. [23] Quick J, Gessler F, Dutzmann S, et al. Benefit of tumor resection for recurrent glioblastoma. J Neurooncol 2014;117:365–72. [24] Grabowski MM, Recinos PF, Nowacki AS, et al. Residual tumor volume versus extent of resection: predictors of survival after surgery for glioblastoma. J Neurosurg 2014:1–9. [25] Schucht P, Knittel S, Slotboom J. et al., 5-ALA complete resections go beyond MR contrast enhancement: shift corrected volumetric analysis of the extent of resection in surgery for glioblastoma. Acta Neurochir (Wien) 2014;156:305–12, discussion 312.
Contents lists available at ScienceDirect
Clinical Neurology and Neurosurgery journal homepage: www.elsevier.com/locate/clineuro
Repeat-surgery at Glioblastoma recurrence, when and why to operate? Genevieve Ening a,b,∗ , Mai Thi Huynh b , Kirsten Schmieder a,b , Christopher Brenke a,b a Department of Neurosurgery, Knappschafts-Krankenhaus Bochum-Langendreer, Ruhr-University of Bochum, In der Schornau 23-25, 44892 Bochum, Germany b Department of Neurosurgery, University Medical Center Mannheim, University of Heidelberg, Theodor-Kutzer-Ufer 1, 68167 Mannheim, Germany
a r t i c l e
i n f o
Article history: Received 14 April 2015 Received in revised form 20 May 2015 Accepted 21 May 2015 Available online 28 May 2015 Keywords: Glioblastoma Survival Recurrence Repeat surgery Complications
a b s t r a c t Objective: Glioblastoma (GB) recurrence is inevitable; guidelines for treatment at disease recurrence are deficient. Clinicians are faced with deciding whom to choose for repeat-surgery. This study analyzes recurrence therapy modalities, investigates characteristics of patients operated on at recurrence and evaluates outcome benefit. Methods: Consecutive adult patients operated on for de novo GB at a single institution from 2006 to 2011 were reviewed. Clinical, radiographic and molecular data of 141 patients diagnosed of recurrent disease were assessed. Reasons for recurrence therapy and therapy modalities were reviewed. Univariate analysis was used to analyze differences in parameters of patients operated on at recurrence and those not. Impact of re-operation on survival was evaluated by the Kaplan-Meier method and Log-rank test. Results: 53 (38%) patients were selected for repeat surgery upon recurrent disease, this was followed by either chemotherapy (CT) (40%), radiotherapy (8%) or both (49%). 57 (40%) patients received CT alone, which was the most frequent mono-second-line therapy opted for. Most frequent indications for repeatsurgery were maximum possible tumor resection mass reduction and symptom relief (62% and 21%, respectively). Univariate analysis of re-operated vs. not operated patients, showed significant differences for age (p = 0.0001* ) and Karnofsky Performance status (KPS) >70 at both primary and repeat tumor resection (p = 0.013* and 0.0001* , respectively). The operated group had a significantly lower Charlsoncomorbidity-index ≤ 3 (p = 0.004* ) and larger tumor size (p = 0.0001* ). Complication risk at recurrence was not significantly different between groups (p = 0.069). However, patients chosen for repeat surgery had significantly less complications at index surgery (p = 0.006* ). Median time from recurrence to death was 11 months (range, 1–33 months) for operated patients as opposed to 5 months (range, 0–22 months) for not operated patients. The former survived significantly longer; 19 months compared to 13 months for those not operated upon (p = 0.002* ). Conclusions: Our study depicts that patients eligible for repeat-surgery at GB recurrence are characterized by a KPS > 70% before primary and repeat-surgery, Charlson-comorbidity-index ≤ 3, large tumor size and young age. These well-selected patients survive significantly longer after repeat-surgery without being at a higher complication risk in comparison to patients not operated upon. © 2015 Published by Elsevier B.V.
1. Introduction Recurrent disease in Glioblastoma (GB) patients is almost ubiquitous irrespective of improved therapy for newly diagnosed GB [1]. Over the past decade, treatment strategies for recurrent patients
∗ Corresponding author at: Department of Neurosurgery, Ruhr-University Bochum, In der Schornau 23-25, D-44892 Bochum, Germany. Tel.: +49 23429980255; fax: +49 2342993609. E-mail addresses: [email protected], [email protected] (G. Ening). http://dx.doi.org/10.1016/j.clineuro.2015.05.024 0303-8467/© 2015 Published by Elsevier B.V.
have become more aggressive, patients are offered various options including repeat-surgery, re-radiation, salvage chemotherapy [2] and alternating electric fields, so called tumor-treating fields [3]. Although some authors report of improved survival after repeatsurgery of recurrent tumor [4], the indications for, and efficacy of repeated surgery at time of tumor recurrence are not well defined [5]. In general, evidence based guidelines for treatment decision-upon disease recurrence are lacking [1]. Deciding to opt for repeated surgery at recurrence poses a dilemma for physicians, patients and their families. We reviewed consecutive patients with GB and sought to answer following questions:
90
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
What therapy modalities were decided on at recurrence? When was second-line therapy performed? What were the reasons for or against repeat-surgery? Which patient, tumor and first-line treatment characteristics differentiated the patients re-operated on from those not operated on at tumor recurrence? How does repeat-surgery impact patient outcome?
2. Methods 2.1. Patient selection Between 2006 and 2011 patients >18 years diagnosed of GB at a single academic institution (university hospital Mannheim, Germany) were identified by retrospective review of the neurosurgical department’s database. Institutional review board approval was obtained for all aspects of the study and all patients consented to their medical records being accessed for the study. Tumor diagnosis was determined by a senior neuropathologist according to the WHO classification of central nervous system tumors [6]. Molecular analysis was performed in the same neuropathological department. Tumor recurrence was defined as reappearance in the contrast enhanced magnetic resonance image (MRI) according to the Revised Assessment in Neuro-Oncology (RANO) criteria [7].
2.2. Recorded variables The type of second-line therapy employed; re-surgery, chemotherapy (CT) radiotherapy (RT) or combination of the various modalities was recorded. Furthermore, the times at which therapy was begun and reasons for deciding for or against repeat surgery were reviewed. Demographic parameters included age and gender. Patient variables comprised preoperative functional status, indexed by the Karnofsky Performance status (KPS) and dichotomized as >70% or ≤70%. Preoperative comorbidity at primary tumor surgery was classified by the Charlson-comorbidity index [8] as described elsewhere [8,9] and was further dichotomized as >3 or ≤3. Contrast enhanced MRIs were obtained preoperatively, <48 h after tumor operation (at both primary and repeat surgery), and during follow-up. Two independent reviewers blinded to patient outcomes carried out assessments. Tumor characteristics documented included anatomical tumor location as described previously [10]. In summary, multifocal lesions were defined as multiple separate enhancing lesions with or without connection. A deep lesion was defined as a lesion located in the basal ganglia, the thalamus, the corpus callosum, and/or the brainstem. Eloquent brain was stated as the motor, sensory, speech, or visual cortex, the basal ganglia/internal capsule, the thalamus, or the brainstem. Volumetric analysis including the contrast-enhancing preoperative tumor volume (CE-PTV) and the extent of resection (EOR) were measured as described elsewhere [11], accordingly gross total resection (GTR), subtotal resection (STR) and partial resection (PR) was defined as 100%, >95% <100% or <95% tumor resection, respectively. Molecular analysis was carried out according to standard methods for DNA extraction from formalin-fixed paraffin-embedded tumor tissue or from unfixed frozen tumor tissue samples. O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation analysis was carried out by methylation-specific polymerase chain reaction. Isocitrate dehydrogenase (IDH)-1 codon 132 was analyzed by direct sequencing. In cases of ambiguous results, the IDH1 sequences were amplified by a different set of primers [12,13].
Fig. 1. Bar chart illustrating the different therapy modalities employed at recurrence and the frequency of implementation.
All recorded complications were categorized into neurological (N), regional/surgical (S) and systemic/medical (M) complications, as suggested for patients undergoing craniotomy by the Glioma Outcome Project [14]. Patients were followed up clinically and by MRI every 3 months. Vital status (alive or dead) and time of death was obtained through patient records and information from relatives or treating physicians. Death data was updated through to June 2012. Primary study endpoints were overall survival (OS) from the date of surgery of primary tumor and time to progress after repeat surgery or therapy at recurrence. In cases where death could not be confirmed by any means, the patients were classified as lost to follow-up at the time of their last clinic visit. 2.3. Statistical analysis Statistical analyses were performed with SPSS version 22.0 (SPSS Inc., Chicago, Illinois, USA). Descriptive data was presented as means ± SDs for parametric data and medians with ranges for nonparametric data. Intergroup frequencies were compared using the Fisher exact test/Mann-Whitney U-test for categorical variables and the Student t-test for continuous variables. Two-sided p values less than 0.05 were considered statistically significant. Factors independently associated with choice for re-operation were assessed by univariate analysis. The Student t-test was used for continuous variables and the chi-square for categorical variables. The Kaplan–Meier-method and Log-rank analysis was used to assess the impact of repeat-surgery on outcome. 3. Results Of the all patients reviewed, 18 died within 30 days after surgery of the initial tumor. 43 died without having been diagnosed of a recurrent tumor. 31 patients had no reliable data and were removed from the cohort. 141 patients were diagnosed of a recurrent tumor and made up our final study group. Fig. 1 illustrates the various types of therapy modalities employed for the 141 cases. 57 patients (40%) received CT alone, this was the most frequent mono-second-line therapy decided on. 53 (38%) of the patients were re-operated on at tumor recurrence. Histological analysis showed that whilst 43 patients had vital recurrent tumor, 10 patients showed no vital tumor. Hence, were viewed to have pseudoprogression. Also, out of the 53 operated cases, 49% received RT and CT, (32%) CT and (8%) RT alone following repeat-surgery. Only six patients (4%) did not receive further therapy following repeat surgery. This was due to patient refusal in five cases and deteriorated condition following a surgical complication after repeat surgery in one case. Fig. 2 shows the time point at which second-line therapy was implemented.
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
91
Table 1 Summary of clinical, treatment, tumor and imaging characteristics of 141 patients with recurrent glioblastoma and univariate analysis of factors in relation to surgery of recurrent tumor. Parameter
Patients No. (%) Demographics Age (mean +-SD)
Fig. 2. Bar chart depicting the time points of recurrence treatment commence and corresponding number of patients.
In 70 cases, more than 50% of patients’ recurrent tumor was distinctly diagnosed via radiographic assessment and second-line therapy was commenced immediately hereafter. In contrast hereto for 3 cases recurrence was unclear from radiographic imaging therefore time elapsed before second-line therapy initiation. No exact specification for the time point of second-line therapy commence was extractable in 15 cases. In 18 patients CT/RT was performed as first option, upon further radiographic progress reoperation was then decided upon. In 22 cases, repeat-surgery was decided on directly at diagnosis of recurrence. Fig. 3 sums up the reasons for deciding to operate patients at recurrence. In most cases (33) the aim of re-operation was firstly to attain maximum possible tumor resection and reduce tumor mass and secondly to relieve symptoms (11 cases). In 3 cases, initial decision of CT/RT was changed to repeat-surgery as the former therapy modalities were not tolerated by the patients. 38 patients were in a bad clinical condition for surgery and of these best supportive care was offered in 9 cases. As illustrated in Fig. 1 8 patients had recurrent tumor locations not eligible for surgery. The mean time interval between operations was 9 ± 6.5 months. 3.1. Univariate analysis of characteristics Table 1 summarizes demographic, patient, tumor and clinical parameters of the 141 patients with recurrent tumors. 53 (38%) patients underwent surgery and 88 (62%) were treated otherwise Fig. 1. The two groups had similar gender distribution (p = 0.15). Patients operated on were significantly younger mean 55.4 years compared to 63.7 years (p = 0.0001* ). Also, patient characteristics (KPS ≥ 70, CCI < = 3) at index surgery were significantly distinct between groups (p = 0.013* and 0.004* , respectively). This was also the case for KPS ≥ 70, at recurrent tumor operation (p = 0.0001* ). Tumor and molecular characteristics (anatomic and eloquent location, CE-PTV, EOR, MGMT and IDH 1 status at index surgery proved no significant difference between the two groups. Although there was no difference in initial tumor volume (CE-PTV) in the operated and not operated groups, patients who were operated on at recurrence had significantly larger recurrent tumor volumes 28.6 cm3 vs. 15.4 cm3 (p = 0.006* ). Group comparison with regard to first-line therapy; combined radiochemotherapy according to the EORTC-Protocol or other therapy modalities showed no significant difference (p = 0.062). Patients operated on at recurrence had significantly less complications at index surgery compared to those not operated (p = 0.006* ). However, there was no significant difference in recurrence treatment compliation rate between those chosen for repeat surgery and those who were not (p = 0.069).
No repeatsurgery
Repeatsurgery
88 (62)
53 (38)
63.7
55.4
Univariate analysis p Value
0.0001* 0.51
Sex Male Female
47 (62) 41 (63)
29 (38) 24 (37)
Patient characteristics KPS at index surgery <70 >70
18 (86) 70 (58)
3 (14) 50 (42)
KPS at tumor recurrence <70 >70
44 (86) 34 (46)
7 (14) 40 (54)
CCI at index surgery ≤3 >3
42 (53) 46 (75)
38 (47) 15 (25)
0.013*
0.0001*
0.004*
Tumor Characteristics Anatomical tumor localization at index surgery 25 (57) Frontal Temporal 21 (60) 11 (61) Parietal Occipital 2 (40) 26 (72) Not localized to one site 3 (100) Others
19 (43) 14 (40) 7 (39) 3 (60) 10 (28) 0 (0)
Eloquent tumor localization at index surgery 12 (67) Eloquent 50 (70) Near-Eloquent Non-Eloquent 26 (51)
6 (33) 22 (31) 25 (49)
0.411
0.106
Tumor volume CE-PTV after first operation CE-PTV (cm3 ) 33.13
36.48
0.573
Tumor volume CE-PTV at tumor recurrence 15.37 CE-PTV (cm3 )
28.57
0.016*
EOR initial surgery <95% >95% < 100% 100%
22 (56) 20 (54) 22 (69)
17 (44) 17 (46) 10 (31)
MGMT Methylated Unmethylated
26 (54) 36 (61)
22 (46) 23 (39)
IDH1 Mutant Wildetype
2 (33) 66 (64)
4 (67) 38 (36)
Clinical characteristics Complication after index surgery Yes No
60 (52) 28 (49)
24 (32) 29 (51)
Complication after recurrent operation 23 Yes 65 No
21 32
First-line adjuvant therapy RT/TMZ → TMZ Other
0.419
0.302
0.149
0.006*
0.069
0.062 66 (59) 22 (86)
46 (41) 7 (14)
Abbreviations: CE-PTV-contrast-enhancing preoperative tumor volume, MGMTO6-methylguanine-DNA methyltransferase, IDH-1-Isocitrate dehydrogenase, EORextent of resection, GTR-gross total resection, STR-subtotal resection, PR partial resection PR, KPS Karnofsky Performance status and CCI Charlson-comorbidityindex, RT/TMZ → TMZ Concomitant Radiochemotherapy with Temozolomide (TMZ) and following TMZ therapy.
92
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
Fig. 3. Bar chart showing the frequencies of reasons for deciding to opt for repeat-surgery in patients.
3.2. Outcome The median time interval from diagnosis of recurrent disease to death was 11 months (range, 1–33 months) for operated patients as opposed to 5 months (range, 0–22 months) for not operated patients. Patients operated on at recurrence survived significantly longer 19 months compared to 13 months for those not operated upon (p = 0.002* ). This is illustrated by the Kaplan-Meier curve in Fig. 4. Evaluation of first-line therapy employed in patients with histological evidence of vital tumor as compared to those diagnosed of pseudoprogression showed that 37 out of the 43 patients with vital recurrent tumor received initial combined radiochemotherapy. In contrast, 9 out of the 10 patients were classified to have a pseudoprogression. Kaplan-Meier outcome analysis showed no difference in outcome with regard to patients with vital recurrent tumor and those with pseudoprogression. 4. Discussion 4.1. Therapy modalities decided on at recurrence. Whereas 91% of patients received some sort of therapy at recurrence, 9% of patients received no tumor specific treatment. In our study, 40% of patients received CT alone. In contrast to these, 38% of patients were chosen for repeat surgery. Of the latter, only 2 patients received surgery as a mono-therapy. This is in accordance with literature, where it is stated that for most patients, CT remains the treatment of choice at recurrence and for patients with a substantially reduced performance status, supportive care measures alone may be more appropriate [1]. Similar to treatment at primary tumor diagnosis, at recurrence, re-surgery was a therapy opted for in combination with other therapy modalities. Previous reports state that surgery as a mono-therapy in GB treatment is not reasonable and has to be combined with other therapies [16]. Hence, repeat-surgery seems a just therapy if further therapy options are available. 4.2. Time point for second-line therapy In 70 cases from our study, recurrent diagnosis was well-defined according to MRI imaging criterion and hence second-line therapy was commenced without delay. Three patients were observed initially and decided on treatment after further MRI scanning. In 22 patients, surgery was the initial second-line therapy option. In contrast, 18 patients received CT/RT at second-line and were
then re-operated upon response failure. This implies that, at tumor recurrence although the decision to offer second-line therapy is clear there is often uncertainty as to which therapy modality to opt for. Although some scores have been defined to aid decision making [16], they are not well established in clinical routine. Similar to most published studies in this field, our review also lacks data on quality of life [2]. In order to account for patients and relatives opinion regarding the time point to commence second-line therapy, this outcome measure may be of importance and should be considered in future studies. 4.3. Reasons for or against repeat-surgery Deciding to operate patients at recurrence was to attain maximum tumor resection allowing tumor mass effect reduction in 62% cases and to achieve symptom relief in 21% cases. Moreover, deteriorated patient condition was most often the reason why surgery was not opted for at recurrence (38 cases). In 10 cases, therapies other than repeat surgery were deemed more appropriate. We meticulously reviewed all protocols of the interdisciplinary neurooncological tumor board, this provided data on the reasons for or against repeat-surgery in each patient. In cases were emergency surgery was decided upon due to mass effect, the information was obtained from the emergency admission protocols. Furthermore, univariate analysis showed that irrespective of the tumor size at index surgery, at recurrence patients with significantly larger tumors were chosen for repeat-surgery. According to literature reports, mass reduction and symptom relief are the main aims for surgery at primary tumor diagnosis [17]. This implies that, reasons for patient repeat-surgery at recurrence are comparable to those for index surgery but still have an independent value. 4.4. Patient, tumor and treatment characteristics for repeat-surgery and impact on outcome Our results show that patients selected for repeat surgery are significantly younger, have a better functional status and less comorbidities. This may reflect some sort of selection bias favoring good candidates for repeat-surgery. To reduce this effect we dichotomized parameters such as KPS and CCI, lowering reviewer subjectivity. The confounding effect of age and performance status was investigated by Stark et al. Their data suggests that re-surgery should be considered in all patients with favorable performance status regardless of age [18]. Also Park et al. in trying to develop a preoperative scoring system, the NIH recurrent GBM scale, considered KPS but not age to be a more relevant parameter [16].
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
93
Fig. 4. Kaplan-Meier curve depicting significant difference in overall survival of the repeat-surgery group median of 19 months compared to not operated group 13 months (p = 0.002* ).
The results of our study clearly show that patients who are reoperated at recurrence survive longer after recurrence (11 months) and have a longer OS of 19 months. This implies an outcome benefit for repeat-surgery. Our findings of a 9 months mean time interval between operations accords with previous results where a time interval of at least 6 months between operations proved to be an important predictor of benefit from reoperation. {Hervey-Jumper, 2014 #3206}Moreover, patients chosen for repeat surgery had had a significantly less complication rate at index surgery (p = 0.006* ), this effect was no more evident at repeat-surgery (p = 0.069). This relates to previous results which state that for recurrent GB, an improvement in overall survival can be attained when the EOR is beyond 80%, emphasizing that this survival benefit must be balanced against the risk of neurological morbidity, which does increase with more aggressive cytoreduction. However, this is only in the early postoperative period [19]. This early postoperative neurological deterioration may be explained by findings from other studies where it was shown that rather than cortical or subcortical structural damage of eloquent brain tissue alone, peri- or postoperative ischemic lesions play a crucial role in the development of surgery-related motor deficits [20]. This suggests that the complication rate at recurrent surgery is irrespective of complications at primary surgery and each surgery has an individual complication risk. Hence, when deemed eligible patients should be chosen for repeat surgery despite the fact that they experienced complications at initial tumor resection. However, to keep complication rate at a minimum, modern strategies of maximal safe resection 5-aminolevulinic acid (ALA), intraoperative monitoring and intraoperative MRI should be employed. Additionally, in defining eloquent brain, areas directly adjacent to the M1 and/or M2 segments of the middle cerebral artery have not traditionally been considered as eloquent brain regions. Hence, injury to the above named vessels during surgery can result in damage to the eloquent brain regions they supply [21]. This difference
in anatomical and functional definition of eloquent regions may explain the insignificant correlations observed in our study as we made only an anatomical distinction of eloquent regions. Future clinical studies have to consider these different definitions of eloquence in order to better reflect the surgical situation. Our findings depict that patients with larger tumors at recurrence measured by volume of contrast enhancing lesion, were more likely to be chosen for repeat surgery and that smaller recurrent tumors were selected for other therapy modalities. This observation was not evident for tumor volumes at index surgery. In contrast, all other tumor characteristics, lobe location, eloquent location and EOR did not differ significantly between groups neither at initial tumor resection nor at recurrence Table 1. These results contradict findings of Bloch et al. and Quick et al. who showed that EOR at recurrence and not at initial surgery impacted survival, suggesting that gross total resection at second craniotomy could overcome the effect of an initial subtotal resection [22,23]. However our findings on CE-PTV somehow relate to reports which state the CE-PTV to be the more significant predictor of survival compared to EOR [24]. Also, it is reported that the resection cavity underestimates the volume of resected tissue and 5-ALA complete resections go significantly beyond the volume of pre-operative contrast-enhancing tumor bulk on MRI [25]. These contradicting volumetric analysis results may be a reflection of the persisting heterogeneity in volumetric measurements employed, and emphasizes the uncertainty and dilemma in defining the value of EOR for both initial and recurrent tumor resections [2]. Hence, future studies will have to consider various pre- and postoperative volumetric definitions in order to allow better comparison of result. 4.5. Study strengths and limitations Our data was obtained from a single institution making up a small cohort size, which in turn minors the statistical analysis.
94
G. Ening et al. / Clinical Neurology and Neurosurgery 136 (2015) 89–94
However, this has the advantage of a single institutional data source reducing possible confounding effects of differences in the access to health care services between health centers. Although limited by the retrospective study design the pseudoprogression rate of 19% in our study cohort correlates with numerous previous reports [15] hence allowing representative analysis and conclusions. In differentiating between recurrence diagnosis and therapy induced pseudoprogression, modern imaging techniques, such as MRSpectroscopy or FET-PET analysis may have been of assistance. Unfortunately at the time of our study these techniques were up coming at our institution and not part of clinical routine. Hence we do not have comprehensive data on this account.” Furthermore, many retrospective studies published lack molecular data correlations [2] one strength of our study is the assessment of MGMT methylation status and IDH1 mutation status in correlation to repeat-surgery at recurrence. However, we did not see any significant correlation between these molecular parameters and decision for repeat-surgery. This may be explained by the fact that the strong predictive effect of MGMT methylation status on outcome is associated with response to CT [5] and not surgical treatment per se. 5. Conclusion In our study patients chosen for repeat-surgery at GB recurrence were of a younger age, showed a KPS > 70% before primary and repeat-surgery, had less comorbidity (CCI ≤ 3) and a larger tumor size compared to not operated patients. Patients selected by these criteria significantly survived longer after repeat-surgery without being at a higher risk for encountering complications. References [1] Preusser M, de Ribaupierre S, Wohrer A, et al. Current concepts and management of glioblastoma. Ann Neurol 2011;70:9–21. [2] Hervey-Jumper SL, Berger MS. Reoperation for recurrent high-grade glioma: a current perspective of the literature. Neurosurgery 2014;75:491–9. [3] Stupp R, Wong ET, Kanner AA, et al. NovoTTF-100A versus physician’s choice chemotherapy in recurrent glioblastoma: a randomised phase III trial of a novel treatment modality. Eur J Cancer 2012;48:2192–202. [4] Helseth R, Helseth E, Johannesen TB, et al. Overall survival, prognostic factors, and repeated surgery in a consecutive series of 516 patients with glioblastoma multiforme. Acta Neurol Scand 2010;122:159–67. [5] Stupp R, Brada M, van den Bent MJ, et al. High-grade glioma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol 2014;25(Suppl. 3), iii93-101. [6] Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 2007;114:97–109.
[7] van den Bent MJ, Wefel JS, Schiff D, et al. Response assessment in neurooncology (a report of the RANO group): assessment of outcome in trials of diffuse low-grade gliomas. Lancet Oncol 2011;12:583–93. [8] Balducci M, Fiorentino A, De Bonis P, et al. Impact of age and co-morbidities in patients with newly diagnosed glioblastoma: a pooled data analysis of three prospective mono-institutional phase II studies. Med Oncol 2012;29: 3478–83. [9] Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis 1987;40:373–83. [10] Lacroix M, Abi-Said D, Fourney DR, et al. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg 2001;95:190–8. [11] Shi WM, Wildrick DM, Sawaya R. Volumetric measurement of brain tumors from MR imaging. J Neurooncol 1998;37:87–93. [12] Hartmann C, Hentschel B, Simon M, et al. Long-term survival in primary glioblastoma with versus without isocitrate dehydrogenase mutations. Clin Cancer Res 2013;19:5146–57. [13] Pusch S, Sahm F, Meyer J, et al. Glioma IDH1 mutation patterns off the beaten track. Neuropathol Appl Neurobiol 2011;37:428–30. [14] Chang SM, Parney IF, McDermott M, et al. Perioperative complications and neurological outcomes of first and second craniotomies among patients enrolled in the Glioma Outcome Project. J Neurosurg 2003;98:1175–81. [15] Brandes AA, Tosoni A, Spagnolli F, et al. Disease progression or pseudoprogression after concomitant radiochemotherapy treatment: pitfalls in neurooncology. Neuro Oncol 2008;10:361–7. [16] Park JK, Hodges T, Arko L, et al. Scale to predict survival after surgery for recurrent glioblastoma multiforme. J Clin Oncol 2010;28:3838–43. [17] Gulati S, Jakola AS, Nerland US, et al. The risk of getting worse: surgically acquired deficits, perioperative complications, and functional outcomes after primary resection of glioblastoma. World Neurosurg 2011;76:572–9. [18] Stark AM, Hedderich J, Held-Feindt J, et al. Glioblastoma—the consequences of advanced patient age on treatment and survival. Neurosurg Rev 2007;30:56–61, discussion 61-52. [19] Oppenlander ME, Wolf AB, Snyder LA, et al. An extent of resection threshold for recurrent glioblastoma and its risk for neurological morbidity. J Neurosurg 2014;120:846–53. [20] Gempt J, Krieg SM, Huttinger S, et al. Postoperative ischemic changes after glioma resection identified by diffusion-weighted magnetic resonance imaging and their association with intraoperative motor evoked potentials. J Neurosurg 2013;119:829–36. [21] Chang EF, Smith JS, Chang SM, et al. Preoperative prognostic classification system for hemispheric low-grade gliomas in adults. J Neurosurg 2008;109:817–24. [22] Bloch O, Han SJ, Cha S, et al. Impact of extent of resection for recurrent glioblastoma on overall survival: clinical article. J Neurosurg 2012;117: 1032–8. [23] Quick J, Gessler F, Dutzmann S, et al. Benefit of tumor resection for recurrent glioblastoma. J Neurooncol 2014;117:365–72. [24] Grabowski MM, Recinos PF, Nowacki AS, et al. Residual tumor volume versus extent of resection: predictors of survival after surgery for glioblastoma. J Neurosurg 2014:1–9. [25] Schucht P, Knittel S, Slotboom J. et al., 5-ALA complete resections go beyond MR contrast enhancement: shift corrected volumetric analysis of the extent of resection in surgery for glioblastoma. Acta Neurochir (Wien) 2014;156:305–12, discussion 312.