- Email: [email protected]
Impact of Laterality on Surgical Outcome of Glioblastoma Patients: A Retrospective Single-Center Study
Accepted Manuscript Impact Of Laterality On Surgical Outcome Of Glioblastoma Patients: A Retrospective Single Center Study Daniel Coluccia, MD, Tabitha Roth, B.Sc, Serge Marbacher, MD, Javier Fandino, MD PII:
S1878-8750(18)30355-3
DOI:
10.1016/j.wneu.2018.02.084
Reference:
WNEU 7499
To appear in:
World Neurosurgery
Received Date: 10 December 2017 Revised Date:
11 February 2018
Accepted Date: 13 February 2018
Please cite this article as: Coluccia D, Roth T, Marbacher S, Fandino J, Impact Of Laterality On Surgical Outcome Of Glioblastoma Patients: A Retrospective Single Center Study, World Neurosurgery (2018), doi: 10.1016/j.wneu.2018.02.084. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT IMPACT OF LATERALITY ON SURGICAL OUTCOME OF GLIOBLASTOMA PATIENTS: A RETROSPECTIVE SINGLE CENTER STUDY
Daniel Coluccia*, MD1; Tabitha Roth*, B.Sc1,2;
Department of Neurosurgery and Brain Tumor Center, Kantonsspital Aarau, Switzerland 2
Department of Health Sciences and Technology, ETH Zurich, Switzerland
Corresponding author: Daniel Coluccia, MD
Kantonsspital Aarau
TE D
Department of Neurosurgery
M AN U
*both authors contributed equally to the study
SC
1
RI PT
Serge Marbacher, MD1; Javier Fandino, MD1
Tellstrasse, 5001 Aarau, Switzerland
Telephone: +41-62-838 4141; Fax: +41-62-838 6629
EP
email: [email protected]; [email protected]
AC C
Keywords. Glioblastoma; laterality; dominant hemisphere; extent of resection
Abbreviations list
5-ALA 5-Aminolevulinic Acid; EOR Extent of Resection; fMRI Functional Magnetic Resonance Imaging; GB Glioblastoma; KPS Karnofsky Performance Status; LH Left Hemisphere, Left Hemisphere GB; MMS Mini Mental State; OS Overall Survival; PFS Progression-Free Survival; RCT Randomized Controlled Trial; RH Right Hemisphere, Right Hemisphere GB;TMZ Temozolomide
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT Abstract Background. Resection of left hemispheric tumors (LH) is often complicated by the risks of causing language dysfunction. Although neurosurgeons’ concerns when operating on the presumed dominant hemisphere are well known, literature evaluating laterality as a predictive
RI PT
surgical parameter in glioblastoma (GB) patients is sparse. We evaluated whether tumor laterality correlated with surgical performance, functional outcome, and survival.
Methods. All GB patients treated at our institution between 2006 and 2016 were reviewed.
and survival in relation to tumor lateralization.
SC
Analysis comprised clinical characteristics, extent-of-resection (EOR), neurological outcome,
M AN U
Results. A total of 235 patients were included. Right hemisphere tumors (RH) were larger and more frequently extended into the frontal lobe. Preoperatively, limb paresis was more frequent in RH, while language deficits were more frequent in LH (p=0.0009 and p<0.0001, respectively). At 6 months after resection, LH patients presented lower Karnofsky-
TE D
Performance-Status (KPS) (p=0.036). More LH patients suffered from dysphasia (p<0.0001), while no difference was seen for paresis. Average EOR was comparable, but complete resection was achieved less often in LH (37.7 vs. 64.8%, p=0.0028). While overall-survival
EP
(OS) did not differ between groups, progression-free-survival (PFS) was shorter in LH (7.4 vs. 10.1 months, p=0.0225).
AC C
Conclusion. LH patients had a pronounced KPS decline and shorter PFS without impact on OS. This observation might partially be attributed to a more conservative surgical resection. Further investigations are needed to assess whether systematic use of awake surgery and intraoperative mapping results in increased EOR as well as in improved quality survival of GB patients.
2
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT Introduction Maximal surgical resection is essential to provide adequate tumor control in glioblastoma (GB) patients 1, 2. Tumor location in or near regions of eloquent neuroanatomy is less amenable to surgery and therefore, benefits of maximal therapy are often counterbalanced
RI PT
by increasing risks of permanent morbidity and decreased quality of life (QoL). Considering the extent of language processing cortex areas and white matter tracts in the frontal, parietal and temporal lobe of the left or dominant hemisphere, options for safe resection itself, as well
SC
as surgical accessibility to invasive lesions are more limited here compared to the opposite hemisphere. However, the impact of tumor laterality on therapy and particularly surgical-
M AN U
related morbidity and outcome in GB patients has not been broadly investigated so far. In the present study, we analyzed the clinical characteristics and patient outcomes of all GB cases treated at our institution in the past 10 years in relation to hemispheric location. We assessed whether tumor lateralization to either hemisphere correlated with extent of resection,
Study Design
EP
Materials and Methods
TE D
functional outcome, and survival.
All adult patients initially diagnosed with GB and consecutively treated at our
AC C
institution between 2006 and 2016 were analyzed. Patients with an infratentorial tumor manifestation as well as those with bilateral location were excluded. As thorough assessment of hemispheric language dominance was not consistently available for every patient, we chose to class for left and right hemisphere, given that in more than 90% of humans’ language function is unilaterally represented on the left side 3. Retrospective data analysis grouped for left- and right-sided GB manifestation comprised clinical characteristics, tumor location and volume, and treatment modality, as well as functional outcome and survival. Location was further subdivided according to tumor extension (e.g., frontal, frontal-parietal, basal ganglia, 3
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT etc.). With respect to treatment modality, we particularly examined the rate of performed resections and biopsy procedures as well as the rate of short-term second surgeries (within one month after the first surgery) aimed at improving the extent of resection when residual tumor was appreciated on postoperative imaging. In addition, the rate of adjuvant
RI PT
radiotherapy and first-line (temozolomide) as well as second-line (bevacizumab)
chemotherapy was registered for each group. Additional salvage therapies after bevacizumab therapy were not specifically registered. Based on CT or MRI scans acquired shortly before
SC
and within one to two days after first-line surgery, tumor volume and extent of resection were calculated using BrainLab ElementsTM software (BrainLab, Feldkirchen, Germany). Tumor
M AN U
volume was defined as the region within the contrast-enhancing formation on pre-and postoperative scans, including necrotic or cystic areas. Extent of resection was considered 100% when complete resection of contrast-enhancing tumor (CREST) was achieved. Postoperative enhancement within the resection cavity was considered residual tumor,
TE D
including linear enhancement. Functional outcome assessment included Karnofsky Performance Scores (KPS) at 6 months as well as documentation of focal neurological deficits 3 to 6 months after surgery as registered by the oncologist in charge. Tumor related
EP
personality disorders and memory dysfunctions were recorded according to reports authored by our neuropsychologists. Overall survival (OS) was defined as the time from initial
AC C
diagnosis until death. Progression free survival (PFS) was defined as the time from initial diagnosis after surgery until the first radiologically confirmed tumor progression and clinical deterioration (RANO criteria) 4. Postoperative MR-imaging was usually obtained every 3 months or at the occurrence of neurological deficits.
Statistical Analysis Survival was assessed using the Kaplan-Meier-Method and tested with the GehanBreslow-Wilcoxon test. Statistical significance level was set to ≤ 5% (p ≤ 0.05) and assessed 4
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT using the two-tailed t test and Fisher’s exact test for continuous variables and discrete variables, respectively. Discrete variables were recorded in mean and standard deviation (mean ± SD). Cox-Proportional hazard models were modeled for OS and PFS, once including all patients and once including all patients with a resection. Data of patients who were lost to
RI PT
follow-up were censured at last contact. Karnofsky Performance Status at 6 months were modeled in a linear model. Since there were outliers with high leverage, a robust model was estimated using MM-type estimators according to Koller and Stahel 5.
Results
M AN U
SC
The study was approved by the local ethics committee (EKNZ Nr. 2017-00038).
From all adult patients diagnosed with a GB (n=281) we excluded 46 (16%) cases due to the above-mentioned exclusion criteria. Patient characteristics of the remaining 235
Tumor Size and Location
TE D
patients are summarized in Table 1.
Analysis of tumor volume at first diagnosis showed a trend towards greater volumes in
EP
RH tumors (30.2 ± 2.1 vs. 35.9 ± 2.6 cm3 in LH and RH, respectively, p = 0.0913). Analysis of specific lobar localization revealed a significantly higher fraction of RH tumors to be
AC C
located (at least in part) within the frontal lobe (43.0% vs. 29.8% for RH and LH respectively, p = 0.0424).
Functional Status Preoperative KPS showed no significant difference between the two groups (81.2 ± 12.7 and 82.6 ± 12.1 in LH and RH respectively, p = 0.4126) (Fig. 1). At presentation, speech and language deficits were more frequent in patients with LH tumors (63.2% vs. 10.0% in LH and RH respectively, p < 0.0001) while limb paresis was documented more often in right5
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT sided tumors (29.8% vs. 51.7% in LH and RH respectively, p = 0.0009). Memory deficits tended to be more frequent in patients with tumors on the left side (34.2 vs. 22.5% in LH and RH respectively, p=0.0587), whereas neglect appeared to be more frequent in patients with RH lesions (5.3% vs. 12.5% in LH and RH respectively, p = 0.0672). Points for Mini Mental
RI PT
Score (MMS) performed at diagnosis were significantly lower in patients with LH tumors (mean 21.9 ± 1.1 vs. 26.3 ± 0.6 for LH and RH respectively, p = 0.0006). (Table 1).
SC
Surgical Management (Table 1)
Surgical management was conducted according to the institutional tumor board
M AN U
consensus, evaluating patients’ clinical state and tumor characteristics with respect to chances for safe tumor removal and adequate postsurgical recovery and oncotherapy. Resection was performed in 177 (75.3%) cases; 55 (23.4%) patients underwent biopsy only, while 3 patients were not operated on. Patients receiving only biopsy were significantly older (mean 67 ± 10.9
TE D
vs. 63 ± 11.3 years, for biopsy and resection respectively, p=0.0227), presented lower mean KPS at diagnosis (79.1 ± 13.5 vs. 83.2 ± 11.3 for biopsy and resection respectively, p=0.0267), and more often suffered from a preoperative paresis (55.6 vs. 36.7% for biopsy
EP
and resection respectively, p = 0.0274). Mean tumor size was smaller in patients with biopsy only (25.7 vs. 35.0 cm3 for biopsy and resection respectively, p = 0.0372) and tumor extended
AC C
more often into the basal ganglia (25.8 vs 3.6% for biopsy and resection respectively, p = 0.0004). No significant differences between left and right hemisphere lesions were detected regarding rate of performed surgical resections (71.1 vs. 79.3% for LH and RH respectively, p = 0.1732) or for percentage of patients receiving only diagnostic biopsy. Frequency of a second surgery for improved extent-of resection (EOR) was similar for both groups (11.4 vs. 13.2% for LH and RH respectively, p = 0.6717). All surgeries were performed using 5-ALA fluorescence guided resection and neuronavigation. Intraoperative MRI (iMRI) to optimize surgical resection was routinely used whenever it was available, finally resulting in equal 6
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT application for LH and RH tumors (40.2 vs. 41.7% in LH and RH, respectively, p = 0.8178). Use of corticospinal tract mapping did not differ between the groups and was applied in overall 14%. At our institution, awake craniotomy is generally indicated for low-grade gliomas in eloquent areas. Only three patients underwent GB resection using awake surgery,
RI PT
all of whom had a tumor in the left dominant hemisphere.
Extent of Resection (Table 2, Fig. 2)
SC
Mean EOR showed no between-group difference (90.9 ± 2.0 and 93.7 ± 1.6 %
respectively for LH and RH, p = 0.2783), as also reflected by similar residual tumor volume.
M AN U
Remarkably, complete resection (100% EOR) was achieved significantly more often in RH tumors (64.8 vs. 37.7% for RH and LH respectively, p = 0.0028).
Adjuvant Therapy (Table 1).
TE D
Standard adjuvant treatment (TMZ plus radiotherapy 6) was applied in 153 (65.1%) of our patients and 16 (6.8%) only underwent radiotherapy. In 41 (17.4%) cases no adjuvant treatment was initiated and 25 (10.6%) patients were lost to follow-up as treatment was
EP
continued at other institutions. The use of TMZ or only radiotherapy did not differ in the two LH and RH groups. The same applied for patients without adjuvant treatment and the use of
AC C
bevacizumab as second-line therapy. Patients undergoing a resection were more likely to receive standard adjuvant therapy (50.0 vs. 80.1% for biopsy and resection, respectively, p < 0.0001).
Functional Outcome (Table 1 and Table 2) Post-surgical deficits were assessed within 3 to 6 months after surgery. More patients with a LH tumor suffered from dysphasia (54.6 vs. 4.4% for LH and RH respectively, p < 0.0001). Newly acquired language deficits after surgery occurred in 9 (5.1%) patients, of 7
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT whom 8 had a LH and only 1 a RH tumor (12.5 vs. 1.6% for LH and RH respectively, p=0.0189). Of these 9 tumors, 3 were located in the parietal lobe, 2 were in the frontotemporal lobe, 2 in the frontal lobe, 1 in the temporal lobe, and 1 in the temporo-occipital lobe. Cases of newly acquired paresis were uniformly distributed to both sides (16.1 vs. 9.8%
RI PT
for LH and RH respectively, p = 0.2996). At 6 months after surgery, RH patients more likely reported to carry on normal daily activities without restrictions (KPS >80, WHO score 0) 7 (36.1 vs. 60.0%, for LH and RH respectively, p=0.011). In patients undergoing resection, RH
SC
patients had a better mean KPS (77.2 ± 15.5 and 83.4 ± 14.81 in LH and RH respectively, p = 0.0357) and significantly more frequently presented a high-performance score (KPS>80) at 6
Survival (Table 2 and Table 3)
M AN U
months (38.0 vs. 64.4 % for LH and RH respectively, p = 0.0073)
Follow-up ended in November 2016 and 177 of our 235 patients were deceased by
TE D
then; one of them committed assisted suicide. Thirty-three patients were still alive; the remaining 25 patients were lost to follow-up. Median survival for all patients was 12.9 months (56 weeks). Patients receiving standard treatment including TMZ and radiotherapy
EP
following resection had a median survival of 18.4 months (80 weeks) as compared to 4.6 months for those who received biopsy only. Kaplan-Meier analysis showed a significant
AC C
difference in PFS between LH and RH patients (32 vs. 44 weeks for LH and RH respectively, p = 0.0225), while the difference of nearly 4 months in median survival between the LH and RH groups did not reach statistical significance (52 vs. 68 weeks for LH and RH respectively, p = 0.2382). To account for selection bias resulting from differences in therapeutic strategies, which particularly affects patients of older age and with low KPS, a separate Kaplan-Meier analysis was performed including only patients who underwent tumor resection. Again, PFS for LH patients was significantly shorter (32 vs. 44 weeks for LH and RH respectively, p = 0.0315). A non-significant median survival difference of 3.2 months (60 vs. 74 weeks for LH 8
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT and RH respectively, p = 0.3518) was detected. Five patients survived for more than 5 years, one of them with a left hemisphere tumor (still alive after 64 months). The remaining four harbored a right hemispheric tumor (123, 96, 70 and 65 months overall survival).
RI PT
Multivariate Analysis
Tumor hemisphere, age at diagnosis, lobar localization, initial tumor size and KPS at diagnosis were tested. Furthermore, time to first radiation, EOR (in 10 units) and tumor size
SC
pre- and post-surgery were included in the models for patients with resections. Each time modeling separates models including either EOR or tumor sizes. Age (HR 1.51 [1.29, 1.75], p
M AN U
< 0.001), temporal localization (HR 1.64 [1.05, 2.56], p = 0.0301) and KPS at diagnosis (HR 0.85 [0.74, 0.97], p = 0.0191) were associated with overall survival in all patients. Age (HR 1.44 [1.14, 1.81], p = 0.0020) and EOR (HR 0.73 [0.60, 0.87], p < 0.001) were predictive factors for OS in patients with a resection. Lateralization to the right (HR 0.71 [0.51, 0.99], p
TE D
= 0.0454), age (HR 1.30 [1.12, 1.52], p < 0.001), temporal localization (HR 1.65 [1.03, 2.65], p = 0.0367), and frontal localization (HR 1.82 [1.07, 3.07], p = 0.0257) were significantly associated with PFS in all patients. Age (HR 1.41 [1.11, 1.80], p = 0.0050), temporal
EP
localization (HR 2.47 [1.20, 5.08], p = 0.0138), occipital localization (HR 2.90 [1.04, 8.09], p = 0.0416), and time to first radiation (HR 1.19 [1.00, 1.41], p = 0.0463) were predictive for
AC C
PFS in patients with a resection. Linear models for KPS at 6 months identified age (p < 0.001) and KPS at diagnosis (p = 0.002) as predictive factors in all patients, whereas hemisphere (p = 0.02), age (p = 0.03), EOR (p < 0.001), and time to first radiation (p = 0.05) were significantly associated with 6-month KPS scores in patients receiving resection.
Discussion We present a retrospective single center study which evaluates the impact of tumor laterality on therapeutic decisions, surgical results, functional outcome, and survival in 235 9
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT consecutively treated GB patients. Our results show that the same therapeutic approach was followed for LH and RH patients, as no difference was found in rates of performed resections or in radio- and chemotherapy treatments. Treatment parameters such as overall EOR, repeated surgery, and compliance to postoperative Stupp regime as recorded in our series are
10
RI PT
well comparable to rates reported from other institutions 8, 9. Consistent with previous reports , OS did not significantly differ between LH and RH patients. There was no difference in
average EOR either; however, complete resection of contrast-enhancing tumor was
SC
significantly less often achieved for LH patients. In addition, PFS was significantly shorter in the LH group. Importantly, language impairments were detected in as much as 54.6% of LH
M AN U
compared to 4.4% of RH patients at 6 months follow-up after resection.
Surgical procedures in the dominant hemisphere would intuitively appear to harbor greater surgical risks for disabling neurological deficits given the relevance of language function, which irrespective of handedness is most commonly represented on the left side (in
TE D
approximately 90% of all patients) 3. Noll et al. showed that patients with left temporal lobe glioma not only suffer from higher neurocognitive impairment preoperatively, they also present a more frequent and severe decline in neurocognitive abilities after surgical resection
EP
compared to patients with right temporal lobe glioma 11, 12. McGirt et al. examined a series of 306 GB patients and associated postoperative language as well as motor deficits with a
AC C
decreased median survival of 3.2 months and 3.6 months, respectively 13. Although the oftenencountered restraint of neurosurgeons towards operating on the left hemisphere is well known 14, literature specifically evaluating laterality as a predictive parameter for surgical results and functional outcome in GB patients is sparse. Further, there is no extensive data available to assess whether complicating neurological deficits caused by surgery on the dominant hemisphere more severely impact the QoL of GB patients and therefore justifies a more conservative resection.
10
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT Polin et al. presented the first well-powered and structured retrospective study assessing the impact of tumor lateralization on survival and functional outcome in 280 GB patients 10. Only age and initial KPS score were identified as significant prognostic variables, while no differences in survival or KPS at follow-up were found between patients with
RI PT
tumors in the dominant or non-dominant hemisphere. The percentages of gross-total-
resections (which was however defined only by the surgeon’s intraoperative judgment) were similar in both groups, while tumors in the non-dominant hemisphere tended to be larger and
SC
more often located in the frontal lobe (consistent with our results). Given the similar
outcomes, Polin et al. suggested that decisions for or against gross-total tumor resection be
M AN U
based only on prognostic factors and practical surgical considerations rather than on laterality. Remarkably, commonly used performance scales to describe neurological function in the course of brain tumor diseases (such as KPS, Modified Rankin Scale, and Barthel Index) do not specifically consider aphasia, which may result in an incomplete reflection of patients’
TE D
QoL as well as inadequate comparisons of outcomes between LH and RH patients. In line with Polin et al., 10 tumor laterality was not significantly associated with OS within our cohort. To account for bias of different surgical strategies within our two groups,
EP
we separately analyzed all patients who underwent resection (n=177). Again, no significant survival difference could be demonstrated (Table 3). These results were confirmed in our
AC C
multivariate analysis. Only age, temporal localization and KPS at diagnosis were associated with OS in all patients, whereas only age and EOR were prognostic in patients with resections. According to our results, we conclude that following contemporary treatment protocols, tumor laterality is not likely to influence OS in GB patients. EOR of at least 7080%, as well as residual tumor volume of less than 5cm3, have been shown to decisively promote GB tumor control and increase OS 1, 15, 16. In line with these previous reports, we confirmed a significant survival benefit from 80% EOR and higher, with 100% resection being significantly superior compared to resections between 90-100% (median survival 72 vs. 11
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT 56 weeks, p= 0.038, data not shown). These results further underline the emerging consensus in the literature that even small tumor residues significantly influence time to progression and thus the QoL of patients. Pushing the boundary of a 100% CREST resection by additional removal of some of the surrounding abnormal FLAIR region might result in additional
RI PT
survival benefit 17. As opposed to OS, time to GB progression was significantly shorter for our LH patients, as also confirmed by multivariate analysis, which identified hemisphere localization as a prognostic factor for PFS (Figure 3, Table 1). High functional outcome as
SC
defined by KPS>80 and mean KPS was also significantly worse for LH patients at 6 months’ follow-up (Figure 1 and Table1,2), and in multivariate analysis, tumor side was significantly
M AN U
associated with KPS at 6 months. We hypothesize that earlier clinical and radiological disease progression in LH patients is at least in part owed to the fact that 100% EOR was achieved significantly less often on the left side (Figure 2). Obviously, the general impact of negative collateral effects of radiotherapy as well as tumor infiltration itself in eloquent neuroanatomy
TE D
must be considered as contributing factors.
The observed differences regarding 100% resection of contrast-enhancing tumor probably reflect our surgeons’ risk-conscious hesitation to aim for aggressive resection in
EP
some LH GBs. Even in left-sided tumors not directly involving critical language processing areas, there are less surgical corridors available, and thus less options for optimal tumor
AC C
exposure compared to tumors analogously located in the right hemisphere. This circumstance may be reflected in the different surgical outcomes. The question arises as to what extent an expanded use of awake craniotomy in critically located GBs would have changed the EOR, PFS and functional outcome of our series. To the best of our knowledge, data specifically analyzing the impact of awake craniotomy on EOR and OS in GB patients has not been published so far, which is why awake craniotomy and brain mapping are still not established as a standard of care for GB resection. In their large series of 1229 GB patients, which included resections using awake craniotomy, Li et al. reported 100% EOR in 71% of their 12
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT patients as compared to our series which overall achieved 100% EOR in only 52% of patients 17
. The reported neurological complications were similar, although comparison of severity is
not feasible. However, the averaged extent of resection of over 90% and overall median survival of 18.4 months seen in our patients who received tumor resection and adjuvant
RI PT
therapy is well comparable to the outcomes reported in Li’s study. With respect to
neurological deficits and QoL in high grade glioma patients, a recent review found no
significant impact of awake surgery 18, although the reported patient numbers were too small
SC
to draw solid conclusions. Recently published data from reference centers have shown that awake surgery can be accomplished even for GB tumors with remarkable safety 19; therefore,
M AN U
further analysis is needed to assess whether this approach increases “resectability”, EOR and survival. More detailed evaluation regarding patients' as well as relatives’ satisfaction and QoL assessment in cases with postsurgical aphasia are further needed to validate to what extent the often-encountered hesitation of neurosurgeons when confronted with left-sided
TE D
GBs is justified.
Our study has several limitations. Being a retrospective single center study, it accordingly comes with all the restrictions, including missing data, selection bias and
EP
interdependence of variables, inherent to such a study. For example, results from our multivariate analysis associating lobar tumor extension with outcome should be treated with
AC C
caution due to strong interdependence of different tumor localizations. In addition, the available information in our records did not include a designated questionnaire to thoroughly assess patients’ QoL during the course of the disease.
Conclusion The results of our study confirm previous reports of a comparable OS between patients with right and left hemisphere GBs. However, it adds new insight in terms of functional outcome and PFS: In our series, LH patients had a significantly shorter time to progression 13
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT and a faster decline in functional abilities, which might at least in part be caused by a more conservative resection achieving significantly less often 100% EOR in these patients. Intensified research is needed to confirm whether systematic implementation of awake
RI PT
surgery results in improved resection grade and quality survival in GB patients.
Figure Legends
SC
Figure 1. Stacked column chart showing distribution of KPS before and 6 months after
surgery. Patients with left-sided GB (L) experienced a more pronounced performance decline compared to right-sided GB (R) (KPS >80 in 36.2 vs. 65.0% for LH and RH GB respectively,
M AN U
p=0.01).
Figure 2. Circle chart displaying EORs achieved for LH and RH GB (A). No significant between-group difference was found for average EOR, which exceeded 90 % in both groups. In a significantly higher percentage of patients with RH GBs, total resection (100%) was achieved (in 64.8% of RH vs. 37.7% of LH, p=0.0028) (B).
(p = 0.0225).
EP
Acknowledgments
TE D
Figure 3. Kaplan Meier plot. Median PFS was 32 and 44 weeks for LH and RH, respectively
Statistical analysis was performed in collaboration with the Department of Clinical Research
AC C
at the University of Basel, Switzerland.
Disclosures
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
Funding This study was not supported by any specific funding or sponsor.
14
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT References
6.
7. 8.
9.
10.
11.
12. 13.
14.
15.
RI PT
SC
5.
M AN U
4.
TE D
3.
EP
2.
Sanai N, Polley MY, McDermott MW, Parsa AT, Berger MS. An extent of resection threshold for newly diagnosed glioblastomas. J Neurosurg. Jul 2011;115(1):3-8. Stummer W, Pichlmeier U, Meinel T, Wiestler OD, Zanella F, Reulen H-J. Fluorescence-guided surgery with 5-aminolevulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial. The Lancet Oncology. 2006;7(5):392-401. Mazoyer B, Zago L, Jobard G, et al. Gaussian mixture modeling of hemispheric lateralization for language in a large sample of healthy individuals balanced for handedness. PLoS One. 2014;9(6):e101165. Wen PY, Macdonald DR, Reardon DA, et al. Updated response assessment criteria for high-grade gliomas: response assessment in neuro-oncology working group. J Clin Oncol. Apr 10 2010;28(11):1963-1972. Koller M, Stahel WA. Sharpening Wald-type inference in robust regression for small samples. Computational Statistics & Data Analysis. 2011;55(8):2504-2515. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. Mar 10 2005;352(10):987996. Schag CC, Heinrich RL, Ganz PA. Karnofsky performance status revisited: reliability, validity, and guidelines. J Clin Oncol. Mar 1984;2(3):187-193. Kreth FW, Thon N, Simon M, et al. Gross total but not incomplete resection of glioblastoma prolongs survival in the era of radiochemotherapy. Ann Oncol. Dec 2013;24(12):3117-3123. Marko NF, Weil RJ, Schroeder JL, Lang FF, Suki D, Sawaya RE. Extent of resection of glioblastoma revisited: personalized survival modeling facilitates more accurate survival prediction and supports a maximum-safe-resection approach to surgery. J Clin Oncol. Mar 10 2014;32(8):774-782. Polin RS, Marko NF, Ammerman MD, et al. Functional outcomes and survival in patients with high-grade gliomas in dominant and nondominant hemispheres. J Neurosurg. Feb 2005;102(2):276-283. Noll KR, Weinberg JS, Ziu M, Benveniste RJ, Suki D, Wefel JS. Neurocognitive Changes Associated With Surgical Resection of Left and Right Temporal Lobe Glioma. Neurosurgery. Nov 2015;77(5):777-785. Noll KR, Ziu M, Weinberg JS, Wefel JS. Neurocognitive functioning in patients with glioma of the left and right temporal lobes. J Neurooncol. Jun 2016;128(2):323-331. McGirt MJ, Mukherjee D, Chaichana KL, Than KD, Weingart JD, Quinones-Hinojosa A. Association of surgically acquired motor and language deficits on overall survival after resection of glioblastoma multiforme. Neurosurgery. Sep 2009;65(3):463-469; discussion 469-470. Neugebauer H, Creutzfeldt CJ, Hemphill JC, 3rd, Heuschmann PU, Juttler E. DESTINYS: attitudes of physicians toward disability and treatment in malignant MCA infarction. Neurocrit Care. Aug 2014;21(1):27-34. Chaichana KL, Jusue-Torres I, Navarro-Ramirez R, et al. Establishing percent resection and residual volume thresholds affecting survival and recurrence for patients with newly diagnosed intracranial glioblastoma. Neuro Oncol. Jan 2014;16(1):113-122.
AC C
1.
15
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT
RI PT
SC M AN U
19.
TE D
18.
EP
17.
Stummer W, Pichlmeier U, Meinel T, et al. Fluorescence-guided surgery with 5aminolevulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial. Lancet Oncol. May 2006;7(5):392-401. Li YM, Suki D, Hess K, Sawaya R. The influence of maximum safe resection of glioblastoma on survival in 1229 patients: Can we do better than gross-total resection? J Neurosurg. Apr 2016;124(4):977-988. Nickel K, Renovanz M, Konig J, et al. The patients' view: impact of the extent of resection, intraoperative imaging, and awake surgery on health-related quality of life in high-grade glioma patients-results of a multicenter cross-sectional study. Neurosurg Rev. Mar 06 2017. Hervey-Jumper SL, Li J, Lau D, et al. Awake craniotomy to maximize glioma resection: methods and technical nuances over a 27-year period. J Neurosurg. Aug 2015;123(2):325-339.
AC C
16.
16
ACCEPTED MANUSCRIPT TABLE 1 – Characteristics, treatment modalities and clinical outcome for left vs. right hemisphere GBs
Left hemisphere
Right hemisphere
(114 patients)
(121 patients)
Men (%)
70 (61.4)
81 (66.9)
Women (%)
44 (38.6)
40 (33.1)
64.2 ± 11.46
63.4 ± 11.46
42 (36.8)
38 (31.4)
81.2 ± 12.7
82.6 ± 12.1
KPS > 80 (%)
52 (45.6)
62 (52.1)
Time symptom onset (weeks)
6.1 ± 8.8
5.5 ± 6.7
Mini Mental State (MMS)
21.9 ± 6.1
26.3 ± 3.3
0.0006
Dysphasia (%)
72 (63.2)
12 (10.0)
<0.0001
Paresis (%)
34 (29.8)
62 (51.7)
0.0009
Neglect (%)
6 (5.3)
15 (12.5)
0.0672
P-Value
Patient Characteristics
Mean initial KPS
0.6149 0.4103 0.4126 0.3597 0.5714
Personality (%)
36 (31.6)
44 (36.7)
0.4908
Memory (%)
39 (34.2)
27 (22.5)
0.0587
Vision (%)
20 (17.5)
24 (20.0)
0.7382
Seizures (%)
29 (25.9)
42 (35.0)
0.1546
TE D
M AN U
Deficits
RI PT
Age ≥ 70y (%)
SC
Mean age (years)
0.4150
34 (29.8)
52 (43.0)
0.0424
48 (42.1)
53 (43.8)
0.8952
18 (15.8)
12 (9.9)
0.2404
40 (35.1)
36 (29.8)
0.4050
5 (4.4)
9 (7.4)
0.4125
30.2
35.9
0.0913
Resection (%)
81 (71.1)
96 (79.3)
0.1732
Biopsy only (%)
32 (28.1)
23 (19.0)
0.1233
No surgery (%)
1 (0.9)
2 (1.7)
>0.9999
Days at hospital (d)
15.0 ± 6.3
16.4 ± 6.8
0.1160
Rehabilitation (%)
45 (40.2)
49 (41.2)
0.8941
Re-surgery for EOR (%)
13 (11.4)
16 (13.2)
0.6717
72 (74.2)
81 (71.7)
0.7562
8 (8.3)
8 (7.1)
0.7985
17 (17.5)
24 (21.2)
0.6009
Tumor Characteristics Tumor Extension Frontal (%) Temporal (%)
Parietal (%) Basal Ganglia (%) 3
AC C
Initial volume (cm )
EP
Occipital (%)
Surgical Management
Adjuvant Therapy TMZ + Radiotherapy (%) Radiotherapy only (%) No adjuvant therapy (%)
ACCEPTED MANUSCRIPT Bevacizumab (%)
31 (39.7)
33 (41.3)
0.8724
39.8 ± 18.5
35.5 ± 9.0
0.0680
Dysphasia post-op (%)
42 (54.6)
3 (4.4)
<0.0001
Paresis post-op (%)
27 (36.0)
24 (35.3)
0.9299
76.38 ± 15.3
80.92 ± 16.9
0.1227
KPS > 80 at 6 months (%)
21 (36.2)
39 (65.0)
0.0113
PFS weeks (95% CI of HR)
32 (0.98-2.0)
44 (0.49-1.0)
OS weeks (95% CI of HR)
52 (0.87-1.6)
68 (0.63-1.15)
Time to first radiation (d)
AC C
EP
TE D
M AN U
SC
Mean KPS at 6 months
RI PT
Outcome
0.0225 0.2382
ACCEPTED MANUSCRIPT TABLE 2 – Extent of Resection, functional outcome and survival analyzed for each hemisphere
Left hemisphere
Right hemisphere
(81 patients)
(96 patients)
Residual Volume (cm3)
2.7 ± 4.7
2.4 ± 5.8
0.7336
Extent of resection (%)
90.9 ± 15.8
93.7 ± 13.6
0.2783
100% EOR (%)
23 (37.7)
46 (64.8)
0.0028
Intraoperative MRI (%)
45 (40.2)
50 (41.7)
0.8178
New Dysphasia (%)
8 (12.5)
1 (1.6)
0.0189
New Paresis (%)
10 (16.1)
6 (9.8)
77.2 ± 15.52
83.39 ± 14.81
0.0357
19 (38.0)
38 (64.4)
0.0073
44 (0.5-1.1)
0.0315
P-Value
Mean KPS at 6 months KPS > 80 at 6 months (%)
32 (0.91-2.0)
OS (weeks) (95% CI of HR)
60 (0.79-1.62)
M AN U
PFS (weeks) (95% CI of HR)
SC
Functional Outcome
RI PT
Resection Characteristics
74* (0.61-1.27)
AC C
EP
TE D
* One patient was excluded from survival statistics due to commitment of assisted suicide
0.2996
0.3518
ACCEPTED MANUSCRIPT TABLE 3 – Survival statistics analyzed for anatomical location and treatment plan
No. of patients (%)
Median Survival (weeks) (95% CI of HR)
All patients
234*
56
Standard Treatment**
130 (56)
80 (0.55-0.89)
P-Value
<0.0001
114 (49)
52
Right hemisphere
120* (51)
68
left
34
52
right
51*
72
left
40
60
right
36
Parietal lobe,
Temporal lobe, left
48
right
53
Occipital lobe, left
18
right
12
Surgical Management Biopsy only
42
56
176* (75)
68 (0.21-0.51)
Left hemisphere
81 (71)
60
Right hemisphere
95* (79)
74
153
80 (0.12-0.68)
16
30 (1.5-8.4)
Radiotherapy only
TE D
TMZ + Radiotherapy
EP
* One patient was excluded from survival statistics due to commitment of assisted suicide ** Including resection, temozolomide and radiotherapy
AC C
0.2422
0.7251
0.8086
36
20 (1.96-4.81)
Adjuvant Therapy
0.3246
60
55 (24)
Resection
0.2382
46
M AN U
Frontal lobe,
SC
Left hemisphere
RI PT
Localization
<0.0001
0.3518
<0.0001
ACCEPTED MANUSCRIPT Article Reference WNEU 7499
EP
TE D
Figure 2: Extent of Resection
M AN U
SC
RI PT
Figure 1
AC C
Figure 3: Progressive Free Survival
ACCEPTED MANUSCRIPT Highlights •
Retrospective single center study of glioblastoma patients investigating surgical results, functional outcome, and survival as a function of tumor laterality Patients with either left or right hemispheric tumors did not differ in average extent of resection and overall survival
RI PT
•
Complete tumor resection was less often achieved in patients with left hemispheric tumors
•
Patients with left hemispheric tumors had a faster decline in Karnofsky Performance
AC C
EP
TE D
M AN U
Score and a shorter progression free survival
SC
•
ACCEPTED MANUSCRIPT Disclosure /Conflict of Interest The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper entitled “Impact of laterality on surgical outcome
AC C
EP
TE D
M AN U
SC
RI PT
of glioblastoma patients”. This study was not supported by any specific funding or sponsor
S1878-8750(18)30355-3
DOI:
10.1016/j.wneu.2018.02.084
Reference:
WNEU 7499
To appear in:
World Neurosurgery
Received Date: 10 December 2017 Revised Date:
11 February 2018
Accepted Date: 13 February 2018
Please cite this article as: Coluccia D, Roth T, Marbacher S, Fandino J, Impact Of Laterality On Surgical Outcome Of Glioblastoma Patients: A Retrospective Single Center Study, World Neurosurgery (2018), doi: 10.1016/j.wneu.2018.02.084. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT IMPACT OF LATERALITY ON SURGICAL OUTCOME OF GLIOBLASTOMA PATIENTS: A RETROSPECTIVE SINGLE CENTER STUDY
Daniel Coluccia*, MD1; Tabitha Roth*, B.Sc1,2;
Department of Neurosurgery and Brain Tumor Center, Kantonsspital Aarau, Switzerland 2
Department of Health Sciences and Technology, ETH Zurich, Switzerland
Corresponding author: Daniel Coluccia, MD
Kantonsspital Aarau
TE D
Department of Neurosurgery
M AN U
*both authors contributed equally to the study
SC
1
RI PT
Serge Marbacher, MD1; Javier Fandino, MD1
Tellstrasse, 5001 Aarau, Switzerland
Telephone: +41-62-838 4141; Fax: +41-62-838 6629
EP
email: [email protected]; [email protected]
AC C
Keywords. Glioblastoma; laterality; dominant hemisphere; extent of resection
Abbreviations list
5-ALA 5-Aminolevulinic Acid; EOR Extent of Resection; fMRI Functional Magnetic Resonance Imaging; GB Glioblastoma; KPS Karnofsky Performance Status; LH Left Hemisphere, Left Hemisphere GB; MMS Mini Mental State; OS Overall Survival; PFS Progression-Free Survival; RCT Randomized Controlled Trial; RH Right Hemisphere, Right Hemisphere GB;TMZ Temozolomide
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT Abstract Background. Resection of left hemispheric tumors (LH) is often complicated by the risks of causing language dysfunction. Although neurosurgeons’ concerns when operating on the presumed dominant hemisphere are well known, literature evaluating laterality as a predictive
RI PT
surgical parameter in glioblastoma (GB) patients is sparse. We evaluated whether tumor laterality correlated with surgical performance, functional outcome, and survival.
Methods. All GB patients treated at our institution between 2006 and 2016 were reviewed.
and survival in relation to tumor lateralization.
SC
Analysis comprised clinical characteristics, extent-of-resection (EOR), neurological outcome,
M AN U
Results. A total of 235 patients were included. Right hemisphere tumors (RH) were larger and more frequently extended into the frontal lobe. Preoperatively, limb paresis was more frequent in RH, while language deficits were more frequent in LH (p=0.0009 and p<0.0001, respectively). At 6 months after resection, LH patients presented lower Karnofsky-
TE D
Performance-Status (KPS) (p=0.036). More LH patients suffered from dysphasia (p<0.0001), while no difference was seen for paresis. Average EOR was comparable, but complete resection was achieved less often in LH (37.7 vs. 64.8%, p=0.0028). While overall-survival
EP
(OS) did not differ between groups, progression-free-survival (PFS) was shorter in LH (7.4 vs. 10.1 months, p=0.0225).
AC C
Conclusion. LH patients had a pronounced KPS decline and shorter PFS without impact on OS. This observation might partially be attributed to a more conservative surgical resection. Further investigations are needed to assess whether systematic use of awake surgery and intraoperative mapping results in increased EOR as well as in improved quality survival of GB patients.
2
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT Introduction Maximal surgical resection is essential to provide adequate tumor control in glioblastoma (GB) patients 1, 2. Tumor location in or near regions of eloquent neuroanatomy is less amenable to surgery and therefore, benefits of maximal therapy are often counterbalanced
RI PT
by increasing risks of permanent morbidity and decreased quality of life (QoL). Considering the extent of language processing cortex areas and white matter tracts in the frontal, parietal and temporal lobe of the left or dominant hemisphere, options for safe resection itself, as well
SC
as surgical accessibility to invasive lesions are more limited here compared to the opposite hemisphere. However, the impact of tumor laterality on therapy and particularly surgical-
M AN U
related morbidity and outcome in GB patients has not been broadly investigated so far. In the present study, we analyzed the clinical characteristics and patient outcomes of all GB cases treated at our institution in the past 10 years in relation to hemispheric location. We assessed whether tumor lateralization to either hemisphere correlated with extent of resection,
Study Design
EP
Materials and Methods
TE D
functional outcome, and survival.
All adult patients initially diagnosed with GB and consecutively treated at our
AC C
institution between 2006 and 2016 were analyzed. Patients with an infratentorial tumor manifestation as well as those with bilateral location were excluded. As thorough assessment of hemispheric language dominance was not consistently available for every patient, we chose to class for left and right hemisphere, given that in more than 90% of humans’ language function is unilaterally represented on the left side 3. Retrospective data analysis grouped for left- and right-sided GB manifestation comprised clinical characteristics, tumor location and volume, and treatment modality, as well as functional outcome and survival. Location was further subdivided according to tumor extension (e.g., frontal, frontal-parietal, basal ganglia, 3
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT etc.). With respect to treatment modality, we particularly examined the rate of performed resections and biopsy procedures as well as the rate of short-term second surgeries (within one month after the first surgery) aimed at improving the extent of resection when residual tumor was appreciated on postoperative imaging. In addition, the rate of adjuvant
RI PT
radiotherapy and first-line (temozolomide) as well as second-line (bevacizumab)
chemotherapy was registered for each group. Additional salvage therapies after bevacizumab therapy were not specifically registered. Based on CT or MRI scans acquired shortly before
SC
and within one to two days after first-line surgery, tumor volume and extent of resection were calculated using BrainLab ElementsTM software (BrainLab, Feldkirchen, Germany). Tumor
M AN U
volume was defined as the region within the contrast-enhancing formation on pre-and postoperative scans, including necrotic or cystic areas. Extent of resection was considered 100% when complete resection of contrast-enhancing tumor (CREST) was achieved. Postoperative enhancement within the resection cavity was considered residual tumor,
TE D
including linear enhancement. Functional outcome assessment included Karnofsky Performance Scores (KPS) at 6 months as well as documentation of focal neurological deficits 3 to 6 months after surgery as registered by the oncologist in charge. Tumor related
EP
personality disorders and memory dysfunctions were recorded according to reports authored by our neuropsychologists. Overall survival (OS) was defined as the time from initial
AC C
diagnosis until death. Progression free survival (PFS) was defined as the time from initial diagnosis after surgery until the first radiologically confirmed tumor progression and clinical deterioration (RANO criteria) 4. Postoperative MR-imaging was usually obtained every 3 months or at the occurrence of neurological deficits.
Statistical Analysis Survival was assessed using the Kaplan-Meier-Method and tested with the GehanBreslow-Wilcoxon test. Statistical significance level was set to ≤ 5% (p ≤ 0.05) and assessed 4
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT using the two-tailed t test and Fisher’s exact test for continuous variables and discrete variables, respectively. Discrete variables were recorded in mean and standard deviation (mean ± SD). Cox-Proportional hazard models were modeled for OS and PFS, once including all patients and once including all patients with a resection. Data of patients who were lost to
RI PT
follow-up were censured at last contact. Karnofsky Performance Status at 6 months were modeled in a linear model. Since there were outliers with high leverage, a robust model was estimated using MM-type estimators according to Koller and Stahel 5.
Results
M AN U
SC
The study was approved by the local ethics committee (EKNZ Nr. 2017-00038).
From all adult patients diagnosed with a GB (n=281) we excluded 46 (16%) cases due to the above-mentioned exclusion criteria. Patient characteristics of the remaining 235
Tumor Size and Location
TE D
patients are summarized in Table 1.
Analysis of tumor volume at first diagnosis showed a trend towards greater volumes in
EP
RH tumors (30.2 ± 2.1 vs. 35.9 ± 2.6 cm3 in LH and RH, respectively, p = 0.0913). Analysis of specific lobar localization revealed a significantly higher fraction of RH tumors to be
AC C
located (at least in part) within the frontal lobe (43.0% vs. 29.8% for RH and LH respectively, p = 0.0424).
Functional Status Preoperative KPS showed no significant difference between the two groups (81.2 ± 12.7 and 82.6 ± 12.1 in LH and RH respectively, p = 0.4126) (Fig. 1). At presentation, speech and language deficits were more frequent in patients with LH tumors (63.2% vs. 10.0% in LH and RH respectively, p < 0.0001) while limb paresis was documented more often in right5
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT sided tumors (29.8% vs. 51.7% in LH and RH respectively, p = 0.0009). Memory deficits tended to be more frequent in patients with tumors on the left side (34.2 vs. 22.5% in LH and RH respectively, p=0.0587), whereas neglect appeared to be more frequent in patients with RH lesions (5.3% vs. 12.5% in LH and RH respectively, p = 0.0672). Points for Mini Mental
RI PT
Score (MMS) performed at diagnosis were significantly lower in patients with LH tumors (mean 21.9 ± 1.1 vs. 26.3 ± 0.6 for LH and RH respectively, p = 0.0006). (Table 1).
SC
Surgical Management (Table 1)
Surgical management was conducted according to the institutional tumor board
M AN U
consensus, evaluating patients’ clinical state and tumor characteristics with respect to chances for safe tumor removal and adequate postsurgical recovery and oncotherapy. Resection was performed in 177 (75.3%) cases; 55 (23.4%) patients underwent biopsy only, while 3 patients were not operated on. Patients receiving only biopsy were significantly older (mean 67 ± 10.9
TE D
vs. 63 ± 11.3 years, for biopsy and resection respectively, p=0.0227), presented lower mean KPS at diagnosis (79.1 ± 13.5 vs. 83.2 ± 11.3 for biopsy and resection respectively, p=0.0267), and more often suffered from a preoperative paresis (55.6 vs. 36.7% for biopsy
EP
and resection respectively, p = 0.0274). Mean tumor size was smaller in patients with biopsy only (25.7 vs. 35.0 cm3 for biopsy and resection respectively, p = 0.0372) and tumor extended
AC C
more often into the basal ganglia (25.8 vs 3.6% for biopsy and resection respectively, p = 0.0004). No significant differences between left and right hemisphere lesions were detected regarding rate of performed surgical resections (71.1 vs. 79.3% for LH and RH respectively, p = 0.1732) or for percentage of patients receiving only diagnostic biopsy. Frequency of a second surgery for improved extent-of resection (EOR) was similar for both groups (11.4 vs. 13.2% for LH and RH respectively, p = 0.6717). All surgeries were performed using 5-ALA fluorescence guided resection and neuronavigation. Intraoperative MRI (iMRI) to optimize surgical resection was routinely used whenever it was available, finally resulting in equal 6
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT application for LH and RH tumors (40.2 vs. 41.7% in LH and RH, respectively, p = 0.8178). Use of corticospinal tract mapping did not differ between the groups and was applied in overall 14%. At our institution, awake craniotomy is generally indicated for low-grade gliomas in eloquent areas. Only three patients underwent GB resection using awake surgery,
RI PT
all of whom had a tumor in the left dominant hemisphere.
Extent of Resection (Table 2, Fig. 2)
SC
Mean EOR showed no between-group difference (90.9 ± 2.0 and 93.7 ± 1.6 %
respectively for LH and RH, p = 0.2783), as also reflected by similar residual tumor volume.
M AN U
Remarkably, complete resection (100% EOR) was achieved significantly more often in RH tumors (64.8 vs. 37.7% for RH and LH respectively, p = 0.0028).
Adjuvant Therapy (Table 1).
TE D
Standard adjuvant treatment (TMZ plus radiotherapy 6) was applied in 153 (65.1%) of our patients and 16 (6.8%) only underwent radiotherapy. In 41 (17.4%) cases no adjuvant treatment was initiated and 25 (10.6%) patients were lost to follow-up as treatment was
EP
continued at other institutions. The use of TMZ or only radiotherapy did not differ in the two LH and RH groups. The same applied for patients without adjuvant treatment and the use of
AC C
bevacizumab as second-line therapy. Patients undergoing a resection were more likely to receive standard adjuvant therapy (50.0 vs. 80.1% for biopsy and resection, respectively, p < 0.0001).
Functional Outcome (Table 1 and Table 2) Post-surgical deficits were assessed within 3 to 6 months after surgery. More patients with a LH tumor suffered from dysphasia (54.6 vs. 4.4% for LH and RH respectively, p < 0.0001). Newly acquired language deficits after surgery occurred in 9 (5.1%) patients, of 7
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT whom 8 had a LH and only 1 a RH tumor (12.5 vs. 1.6% for LH and RH respectively, p=0.0189). Of these 9 tumors, 3 were located in the parietal lobe, 2 were in the frontotemporal lobe, 2 in the frontal lobe, 1 in the temporal lobe, and 1 in the temporo-occipital lobe. Cases of newly acquired paresis were uniformly distributed to both sides (16.1 vs. 9.8%
RI PT
for LH and RH respectively, p = 0.2996). At 6 months after surgery, RH patients more likely reported to carry on normal daily activities without restrictions (KPS >80, WHO score 0) 7 (36.1 vs. 60.0%, for LH and RH respectively, p=0.011). In patients undergoing resection, RH
SC
patients had a better mean KPS (77.2 ± 15.5 and 83.4 ± 14.81 in LH and RH respectively, p = 0.0357) and significantly more frequently presented a high-performance score (KPS>80) at 6
Survival (Table 2 and Table 3)
M AN U
months (38.0 vs. 64.4 % for LH and RH respectively, p = 0.0073)
Follow-up ended in November 2016 and 177 of our 235 patients were deceased by
TE D
then; one of them committed assisted suicide. Thirty-three patients were still alive; the remaining 25 patients were lost to follow-up. Median survival for all patients was 12.9 months (56 weeks). Patients receiving standard treatment including TMZ and radiotherapy
EP
following resection had a median survival of 18.4 months (80 weeks) as compared to 4.6 months for those who received biopsy only. Kaplan-Meier analysis showed a significant
AC C
difference in PFS between LH and RH patients (32 vs. 44 weeks for LH and RH respectively, p = 0.0225), while the difference of nearly 4 months in median survival between the LH and RH groups did not reach statistical significance (52 vs. 68 weeks for LH and RH respectively, p = 0.2382). To account for selection bias resulting from differences in therapeutic strategies, which particularly affects patients of older age and with low KPS, a separate Kaplan-Meier analysis was performed including only patients who underwent tumor resection. Again, PFS for LH patients was significantly shorter (32 vs. 44 weeks for LH and RH respectively, p = 0.0315). A non-significant median survival difference of 3.2 months (60 vs. 74 weeks for LH 8
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT and RH respectively, p = 0.3518) was detected. Five patients survived for more than 5 years, one of them with a left hemisphere tumor (still alive after 64 months). The remaining four harbored a right hemispheric tumor (123, 96, 70 and 65 months overall survival).
RI PT
Multivariate Analysis
Tumor hemisphere, age at diagnosis, lobar localization, initial tumor size and KPS at diagnosis were tested. Furthermore, time to first radiation, EOR (in 10 units) and tumor size
SC
pre- and post-surgery were included in the models for patients with resections. Each time modeling separates models including either EOR or tumor sizes. Age (HR 1.51 [1.29, 1.75], p
M AN U
< 0.001), temporal localization (HR 1.64 [1.05, 2.56], p = 0.0301) and KPS at diagnosis (HR 0.85 [0.74, 0.97], p = 0.0191) were associated with overall survival in all patients. Age (HR 1.44 [1.14, 1.81], p = 0.0020) and EOR (HR 0.73 [0.60, 0.87], p < 0.001) were predictive factors for OS in patients with a resection. Lateralization to the right (HR 0.71 [0.51, 0.99], p
TE D
= 0.0454), age (HR 1.30 [1.12, 1.52], p < 0.001), temporal localization (HR 1.65 [1.03, 2.65], p = 0.0367), and frontal localization (HR 1.82 [1.07, 3.07], p = 0.0257) were significantly associated with PFS in all patients. Age (HR 1.41 [1.11, 1.80], p = 0.0050), temporal
EP
localization (HR 2.47 [1.20, 5.08], p = 0.0138), occipital localization (HR 2.90 [1.04, 8.09], p = 0.0416), and time to first radiation (HR 1.19 [1.00, 1.41], p = 0.0463) were predictive for
AC C
PFS in patients with a resection. Linear models for KPS at 6 months identified age (p < 0.001) and KPS at diagnosis (p = 0.002) as predictive factors in all patients, whereas hemisphere (p = 0.02), age (p = 0.03), EOR (p < 0.001), and time to first radiation (p = 0.05) were significantly associated with 6-month KPS scores in patients receiving resection.
Discussion We present a retrospective single center study which evaluates the impact of tumor laterality on therapeutic decisions, surgical results, functional outcome, and survival in 235 9
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT consecutively treated GB patients. Our results show that the same therapeutic approach was followed for LH and RH patients, as no difference was found in rates of performed resections or in radio- and chemotherapy treatments. Treatment parameters such as overall EOR, repeated surgery, and compliance to postoperative Stupp regime as recorded in our series are
10
RI PT
well comparable to rates reported from other institutions 8, 9. Consistent with previous reports , OS did not significantly differ between LH and RH patients. There was no difference in
average EOR either; however, complete resection of contrast-enhancing tumor was
SC
significantly less often achieved for LH patients. In addition, PFS was significantly shorter in the LH group. Importantly, language impairments were detected in as much as 54.6% of LH
M AN U
compared to 4.4% of RH patients at 6 months follow-up after resection.
Surgical procedures in the dominant hemisphere would intuitively appear to harbor greater surgical risks for disabling neurological deficits given the relevance of language function, which irrespective of handedness is most commonly represented on the left side (in
TE D
approximately 90% of all patients) 3. Noll et al. showed that patients with left temporal lobe glioma not only suffer from higher neurocognitive impairment preoperatively, they also present a more frequent and severe decline in neurocognitive abilities after surgical resection
EP
compared to patients with right temporal lobe glioma 11, 12. McGirt et al. examined a series of 306 GB patients and associated postoperative language as well as motor deficits with a
AC C
decreased median survival of 3.2 months and 3.6 months, respectively 13. Although the oftenencountered restraint of neurosurgeons towards operating on the left hemisphere is well known 14, literature specifically evaluating laterality as a predictive parameter for surgical results and functional outcome in GB patients is sparse. Further, there is no extensive data available to assess whether complicating neurological deficits caused by surgery on the dominant hemisphere more severely impact the QoL of GB patients and therefore justifies a more conservative resection.
10
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT Polin et al. presented the first well-powered and structured retrospective study assessing the impact of tumor lateralization on survival and functional outcome in 280 GB patients 10. Only age and initial KPS score were identified as significant prognostic variables, while no differences in survival or KPS at follow-up were found between patients with
RI PT
tumors in the dominant or non-dominant hemisphere. The percentages of gross-total-
resections (which was however defined only by the surgeon’s intraoperative judgment) were similar in both groups, while tumors in the non-dominant hemisphere tended to be larger and
SC
more often located in the frontal lobe (consistent with our results). Given the similar
outcomes, Polin et al. suggested that decisions for or against gross-total tumor resection be
M AN U
based only on prognostic factors and practical surgical considerations rather than on laterality. Remarkably, commonly used performance scales to describe neurological function in the course of brain tumor diseases (such as KPS, Modified Rankin Scale, and Barthel Index) do not specifically consider aphasia, which may result in an incomplete reflection of patients’
TE D
QoL as well as inadequate comparisons of outcomes between LH and RH patients. In line with Polin et al., 10 tumor laterality was not significantly associated with OS within our cohort. To account for bias of different surgical strategies within our two groups,
EP
we separately analyzed all patients who underwent resection (n=177). Again, no significant survival difference could be demonstrated (Table 3). These results were confirmed in our
AC C
multivariate analysis. Only age, temporal localization and KPS at diagnosis were associated with OS in all patients, whereas only age and EOR were prognostic in patients with resections. According to our results, we conclude that following contemporary treatment protocols, tumor laterality is not likely to influence OS in GB patients. EOR of at least 7080%, as well as residual tumor volume of less than 5cm3, have been shown to decisively promote GB tumor control and increase OS 1, 15, 16. In line with these previous reports, we confirmed a significant survival benefit from 80% EOR and higher, with 100% resection being significantly superior compared to resections between 90-100% (median survival 72 vs. 11
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT 56 weeks, p= 0.038, data not shown). These results further underline the emerging consensus in the literature that even small tumor residues significantly influence time to progression and thus the QoL of patients. Pushing the boundary of a 100% CREST resection by additional removal of some of the surrounding abnormal FLAIR region might result in additional
RI PT
survival benefit 17. As opposed to OS, time to GB progression was significantly shorter for our LH patients, as also confirmed by multivariate analysis, which identified hemisphere localization as a prognostic factor for PFS (Figure 3, Table 1). High functional outcome as
SC
defined by KPS>80 and mean KPS was also significantly worse for LH patients at 6 months’ follow-up (Figure 1 and Table1,2), and in multivariate analysis, tumor side was significantly
M AN U
associated with KPS at 6 months. We hypothesize that earlier clinical and radiological disease progression in LH patients is at least in part owed to the fact that 100% EOR was achieved significantly less often on the left side (Figure 2). Obviously, the general impact of negative collateral effects of radiotherapy as well as tumor infiltration itself in eloquent neuroanatomy
TE D
must be considered as contributing factors.
The observed differences regarding 100% resection of contrast-enhancing tumor probably reflect our surgeons’ risk-conscious hesitation to aim for aggressive resection in
EP
some LH GBs. Even in left-sided tumors not directly involving critical language processing areas, there are less surgical corridors available, and thus less options for optimal tumor
AC C
exposure compared to tumors analogously located in the right hemisphere. This circumstance may be reflected in the different surgical outcomes. The question arises as to what extent an expanded use of awake craniotomy in critically located GBs would have changed the EOR, PFS and functional outcome of our series. To the best of our knowledge, data specifically analyzing the impact of awake craniotomy on EOR and OS in GB patients has not been published so far, which is why awake craniotomy and brain mapping are still not established as a standard of care for GB resection. In their large series of 1229 GB patients, which included resections using awake craniotomy, Li et al. reported 100% EOR in 71% of their 12
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT patients as compared to our series which overall achieved 100% EOR in only 52% of patients 17
. The reported neurological complications were similar, although comparison of severity is
not feasible. However, the averaged extent of resection of over 90% and overall median survival of 18.4 months seen in our patients who received tumor resection and adjuvant
RI PT
therapy is well comparable to the outcomes reported in Li’s study. With respect to
neurological deficits and QoL in high grade glioma patients, a recent review found no
significant impact of awake surgery 18, although the reported patient numbers were too small
SC
to draw solid conclusions. Recently published data from reference centers have shown that awake surgery can be accomplished even for GB tumors with remarkable safety 19; therefore,
M AN U
further analysis is needed to assess whether this approach increases “resectability”, EOR and survival. More detailed evaluation regarding patients' as well as relatives’ satisfaction and QoL assessment in cases with postsurgical aphasia are further needed to validate to what extent the often-encountered hesitation of neurosurgeons when confronted with left-sided
TE D
GBs is justified.
Our study has several limitations. Being a retrospective single center study, it accordingly comes with all the restrictions, including missing data, selection bias and
EP
interdependence of variables, inherent to such a study. For example, results from our multivariate analysis associating lobar tumor extension with outcome should be treated with
AC C
caution due to strong interdependence of different tumor localizations. In addition, the available information in our records did not include a designated questionnaire to thoroughly assess patients’ QoL during the course of the disease.
Conclusion The results of our study confirm previous reports of a comparable OS between patients with right and left hemisphere GBs. However, it adds new insight in terms of functional outcome and PFS: In our series, LH patients had a significantly shorter time to progression 13
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT and a faster decline in functional abilities, which might at least in part be caused by a more conservative resection achieving significantly less often 100% EOR in these patients. Intensified research is needed to confirm whether systematic implementation of awake
RI PT
surgery results in improved resection grade and quality survival in GB patients.
Figure Legends
SC
Figure 1. Stacked column chart showing distribution of KPS before and 6 months after
surgery. Patients with left-sided GB (L) experienced a more pronounced performance decline compared to right-sided GB (R) (KPS >80 in 36.2 vs. 65.0% for LH and RH GB respectively,
M AN U
p=0.01).
Figure 2. Circle chart displaying EORs achieved for LH and RH GB (A). No significant between-group difference was found for average EOR, which exceeded 90 % in both groups. In a significantly higher percentage of patients with RH GBs, total resection (100%) was achieved (in 64.8% of RH vs. 37.7% of LH, p=0.0028) (B).
(p = 0.0225).
EP
Acknowledgments
TE D
Figure 3. Kaplan Meier plot. Median PFS was 32 and 44 weeks for LH and RH, respectively
Statistical analysis was performed in collaboration with the Department of Clinical Research
AC C
at the University of Basel, Switzerland.
Disclosures
The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper.
Funding This study was not supported by any specific funding or sponsor.
14
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT References
6.
7. 8.
9.
10.
11.
12. 13.
14.
15.
RI PT
SC
5.
M AN U
4.
TE D
3.
EP
2.
Sanai N, Polley MY, McDermott MW, Parsa AT, Berger MS. An extent of resection threshold for newly diagnosed glioblastomas. J Neurosurg. Jul 2011;115(1):3-8. Stummer W, Pichlmeier U, Meinel T, Wiestler OD, Zanella F, Reulen H-J. Fluorescence-guided surgery with 5-aminolevulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial. The Lancet Oncology. 2006;7(5):392-401. Mazoyer B, Zago L, Jobard G, et al. Gaussian mixture modeling of hemispheric lateralization for language in a large sample of healthy individuals balanced for handedness. PLoS One. 2014;9(6):e101165. Wen PY, Macdonald DR, Reardon DA, et al. Updated response assessment criteria for high-grade gliomas: response assessment in neuro-oncology working group. J Clin Oncol. Apr 10 2010;28(11):1963-1972. Koller M, Stahel WA. Sharpening Wald-type inference in robust regression for small samples. Computational Statistics & Data Analysis. 2011;55(8):2504-2515. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. Mar 10 2005;352(10):987996. Schag CC, Heinrich RL, Ganz PA. Karnofsky performance status revisited: reliability, validity, and guidelines. J Clin Oncol. Mar 1984;2(3):187-193. Kreth FW, Thon N, Simon M, et al. Gross total but not incomplete resection of glioblastoma prolongs survival in the era of radiochemotherapy. Ann Oncol. Dec 2013;24(12):3117-3123. Marko NF, Weil RJ, Schroeder JL, Lang FF, Suki D, Sawaya RE. Extent of resection of glioblastoma revisited: personalized survival modeling facilitates more accurate survival prediction and supports a maximum-safe-resection approach to surgery. J Clin Oncol. Mar 10 2014;32(8):774-782. Polin RS, Marko NF, Ammerman MD, et al. Functional outcomes and survival in patients with high-grade gliomas in dominant and nondominant hemispheres. J Neurosurg. Feb 2005;102(2):276-283. Noll KR, Weinberg JS, Ziu M, Benveniste RJ, Suki D, Wefel JS. Neurocognitive Changes Associated With Surgical Resection of Left and Right Temporal Lobe Glioma. Neurosurgery. Nov 2015;77(5):777-785. Noll KR, Ziu M, Weinberg JS, Wefel JS. Neurocognitive functioning in patients with glioma of the left and right temporal lobes. J Neurooncol. Jun 2016;128(2):323-331. McGirt MJ, Mukherjee D, Chaichana KL, Than KD, Weingart JD, Quinones-Hinojosa A. Association of surgically acquired motor and language deficits on overall survival after resection of glioblastoma multiforme. Neurosurgery. Sep 2009;65(3):463-469; discussion 469-470. Neugebauer H, Creutzfeldt CJ, Hemphill JC, 3rd, Heuschmann PU, Juttler E. DESTINYS: attitudes of physicians toward disability and treatment in malignant MCA infarction. Neurocrit Care. Aug 2014;21(1):27-34. Chaichana KL, Jusue-Torres I, Navarro-Ramirez R, et al. Establishing percent resection and residual volume thresholds affecting survival and recurrence for patients with newly diagnosed intracranial glioblastoma. Neuro Oncol. Jan 2014;16(1):113-122.
AC C
1.
15
GLIOBLASTOMA LATERALITY
ACCEPTED MANUSCRIPT
RI PT
SC M AN U
19.
TE D
18.
EP
17.
Stummer W, Pichlmeier U, Meinel T, et al. Fluorescence-guided surgery with 5aminolevulinic acid for resection of malignant glioma: a randomised controlled multicentre phase III trial. Lancet Oncol. May 2006;7(5):392-401. Li YM, Suki D, Hess K, Sawaya R. The influence of maximum safe resection of glioblastoma on survival in 1229 patients: Can we do better than gross-total resection? J Neurosurg. Apr 2016;124(4):977-988. Nickel K, Renovanz M, Konig J, et al. The patients' view: impact of the extent of resection, intraoperative imaging, and awake surgery on health-related quality of life in high-grade glioma patients-results of a multicenter cross-sectional study. Neurosurg Rev. Mar 06 2017. Hervey-Jumper SL, Li J, Lau D, et al. Awake craniotomy to maximize glioma resection: methods and technical nuances over a 27-year period. J Neurosurg. Aug 2015;123(2):325-339.
AC C
16.
16
ACCEPTED MANUSCRIPT TABLE 1 – Characteristics, treatment modalities and clinical outcome for left vs. right hemisphere GBs
Left hemisphere
Right hemisphere
(114 patients)
(121 patients)
Men (%)
70 (61.4)
81 (66.9)
Women (%)
44 (38.6)
40 (33.1)
64.2 ± 11.46
63.4 ± 11.46
42 (36.8)
38 (31.4)
81.2 ± 12.7
82.6 ± 12.1
KPS > 80 (%)
52 (45.6)
62 (52.1)
Time symptom onset (weeks)
6.1 ± 8.8
5.5 ± 6.7
Mini Mental State (MMS)
21.9 ± 6.1
26.3 ± 3.3
0.0006
Dysphasia (%)
72 (63.2)
12 (10.0)
<0.0001
Paresis (%)
34 (29.8)
62 (51.7)
0.0009
Neglect (%)
6 (5.3)
15 (12.5)
0.0672
P-Value
Patient Characteristics
Mean initial KPS
0.6149 0.4103 0.4126 0.3597 0.5714
Personality (%)
36 (31.6)
44 (36.7)
0.4908
Memory (%)
39 (34.2)
27 (22.5)
0.0587
Vision (%)
20 (17.5)
24 (20.0)
0.7382
Seizures (%)
29 (25.9)
42 (35.0)
0.1546
TE D
M AN U
Deficits
RI PT
Age ≥ 70y (%)
SC
Mean age (years)
0.4150
34 (29.8)
52 (43.0)
0.0424
48 (42.1)
53 (43.8)
0.8952
18 (15.8)
12 (9.9)
0.2404
40 (35.1)
36 (29.8)
0.4050
5 (4.4)
9 (7.4)
0.4125
30.2
35.9
0.0913
Resection (%)
81 (71.1)
96 (79.3)
0.1732
Biopsy only (%)
32 (28.1)
23 (19.0)
0.1233
No surgery (%)
1 (0.9)
2 (1.7)
>0.9999
Days at hospital (d)
15.0 ± 6.3
16.4 ± 6.8
0.1160
Rehabilitation (%)
45 (40.2)
49 (41.2)
0.8941
Re-surgery for EOR (%)
13 (11.4)
16 (13.2)
0.6717
72 (74.2)
81 (71.7)
0.7562
8 (8.3)
8 (7.1)
0.7985
17 (17.5)
24 (21.2)
0.6009
Tumor Characteristics Tumor Extension Frontal (%) Temporal (%)
Parietal (%) Basal Ganglia (%) 3
AC C
Initial volume (cm )
EP
Occipital (%)
Surgical Management
Adjuvant Therapy TMZ + Radiotherapy (%) Radiotherapy only (%) No adjuvant therapy (%)
ACCEPTED MANUSCRIPT Bevacizumab (%)
31 (39.7)
33 (41.3)
0.8724
39.8 ± 18.5
35.5 ± 9.0
0.0680
Dysphasia post-op (%)
42 (54.6)
3 (4.4)
<0.0001
Paresis post-op (%)
27 (36.0)
24 (35.3)
0.9299
76.38 ± 15.3
80.92 ± 16.9
0.1227
KPS > 80 at 6 months (%)
21 (36.2)
39 (65.0)
0.0113
PFS weeks (95% CI of HR)
32 (0.98-2.0)
44 (0.49-1.0)
OS weeks (95% CI of HR)
52 (0.87-1.6)
68 (0.63-1.15)
Time to first radiation (d)
AC C
EP
TE D
M AN U
SC
Mean KPS at 6 months
RI PT
Outcome
0.0225 0.2382
ACCEPTED MANUSCRIPT TABLE 2 – Extent of Resection, functional outcome and survival analyzed for each hemisphere
Left hemisphere
Right hemisphere
(81 patients)
(96 patients)
Residual Volume (cm3)
2.7 ± 4.7
2.4 ± 5.8
0.7336
Extent of resection (%)
90.9 ± 15.8
93.7 ± 13.6
0.2783
100% EOR (%)
23 (37.7)
46 (64.8)
0.0028
Intraoperative MRI (%)
45 (40.2)
50 (41.7)
0.8178
New Dysphasia (%)
8 (12.5)
1 (1.6)
0.0189
New Paresis (%)
10 (16.1)
6 (9.8)
77.2 ± 15.52
83.39 ± 14.81
0.0357
19 (38.0)
38 (64.4)
0.0073
44 (0.5-1.1)
0.0315
P-Value
Mean KPS at 6 months KPS > 80 at 6 months (%)
32 (0.91-2.0)
OS (weeks) (95% CI of HR)
60 (0.79-1.62)
M AN U
PFS (weeks) (95% CI of HR)
SC
Functional Outcome
RI PT
Resection Characteristics
74* (0.61-1.27)
AC C
EP
TE D
* One patient was excluded from survival statistics due to commitment of assisted suicide
0.2996
0.3518
ACCEPTED MANUSCRIPT TABLE 3 – Survival statistics analyzed for anatomical location and treatment plan
No. of patients (%)
Median Survival (weeks) (95% CI of HR)
All patients
234*
56
Standard Treatment**
130 (56)
80 (0.55-0.89)
P-Value
<0.0001
114 (49)
52
Right hemisphere
120* (51)
68
left
34
52
right
51*
72
left
40
60
right
36
Parietal lobe,
Temporal lobe, left
48
right
53
Occipital lobe, left
18
right
12
Surgical Management Biopsy only
42
56
176* (75)
68 (0.21-0.51)
Left hemisphere
81 (71)
60
Right hemisphere
95* (79)
74
153
80 (0.12-0.68)
16
30 (1.5-8.4)
Radiotherapy only
TE D
TMZ + Radiotherapy
EP
* One patient was excluded from survival statistics due to commitment of assisted suicide ** Including resection, temozolomide and radiotherapy
AC C
0.2422
0.7251
0.8086
36
20 (1.96-4.81)
Adjuvant Therapy
0.3246
60
55 (24)
Resection
0.2382
46
M AN U
Frontal lobe,
SC
Left hemisphere
RI PT
Localization
<0.0001
0.3518
<0.0001
ACCEPTED MANUSCRIPT Article Reference WNEU 7499
EP
TE D
Figure 2: Extent of Resection
M AN U
SC
RI PT
Figure 1
AC C
Figure 3: Progressive Free Survival
ACCEPTED MANUSCRIPT Highlights •
Retrospective single center study of glioblastoma patients investigating surgical results, functional outcome, and survival as a function of tumor laterality Patients with either left or right hemispheric tumors did not differ in average extent of resection and overall survival
RI PT
•
Complete tumor resection was less often achieved in patients with left hemispheric tumors
•
Patients with left hemispheric tumors had a faster decline in Karnofsky Performance
AC C
EP
TE D
M AN U
Score and a shorter progression free survival
SC
•
ACCEPTED MANUSCRIPT Disclosure /Conflict of Interest The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper entitled “Impact of laterality on surgical outcome
AC C
EP
TE D
M AN U
SC
RI PT
of glioblastoma patients”. This study was not supported by any specific funding or sponsor