Superior Efficacy of Gross Total Resection in Anaplastic Astrocytoma Patients Relative to Glioblastoma Patients

Accepted Manuscript Superior efficacy of gross total resection in anaplastic astrocytoma patients relative to glioblastoma patients Jennifer A. Padwal, Xuezhi Dong, Brian R. Hirshman, U. Hoi-Sang, Bob S. Carter, Clark C. Chen PII:

S1878-8750(16)00323-5

DOI:

10.1016/j.wneu.2016.02.078

Reference:

WNEU 3779

To appear in:

World Neurosurgery

Received Date: 4 January 2016 Revised Date:

16 February 2016

Accepted Date: 17 February 2016

Please cite this article as: Padwal JA, Dong X, Hirshman BR, Hoi-Sang U, Carter BS, Chen CC, Superior efficacy of gross total resection in anaplastic astrocytoma patients relative to glioblastoma patients, World Neurosurgery (2016), doi: 10.1016/j.wneu.2016.02.078. 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

Page 1

RI PT

Title: Superior efficacy of gross total resection in anaplastic astrocytoma patients relative to glioblastoma patients

Jennifer A. Padwal, Xuezhi Dong, Brian R. Hirshman, Hoi-Sang U, Bob S. Carter, and Clark C. Chen.

School of Medicine

SC

University of California, San Diego

La Jolla, CA 92093

M AN U

9500 Gilman Drive

Codman Center for Clinical Effectiveness in Surgery Massachusetts General Hospital (MGH) 165 Cambridge Street, Suite 403

TE D

Boston, MA 02114

Division of Neurological Surgery

EP

University of California, San Diego 200 West Arbor Drive #8893

AC C

San Diego, CA 92103

Corresponding Author:

Clark Chen, M.D., Ph.D.,

Division of Neurosurgery University of California, San Diego

ACCEPTED MANUSCRIPT

Page 2 3855 Health Science Drive #0987 La Jolla 92093-0987

RI PT

Phone: 619-246-0674 Fax: 858-822-4715 e-mail: [email protected]

AC C

EP

M AN U

TE D

Conflicts of Interest: None

SC

Funding: None

ACCEPTED MANUSCRIPT

Page 3 INTRODUCTION

RI PT

Anaplastic astrocytoma (AA) is a rare form of brain cancer, accounting for 5.9% of primary CNS gliomas1 and with an incidence rate of 0.25 cases per 100,000 persons per year.2 Because of its rarity, clinical studies often group AA and a related but more aggressive form of brain cancer, glioblastoma. Together, these brain cancers are referred to as high-grade gliomas (HGGs).3,4 The

SC

main rationale for this grouping is that AAs inevitably progress to glioblastomas.5 As such,

M AN U

proponents of the nomenclature argue that the two diseases share a common pathophysiology6 and should exhibit comparable responses to therapeutics.7

Recent molecular characterization of AAs, however, suggests fundamental histopathological differences between AAs and glioblastomas. For instance, the available evidence suggests that

TE D

brain tumors harboring Isocitrate Dehydrogenase (IDH) mutations exhibit a distinct epigenetic landscape,8,9 molecular physiology,10,11 and differential response to surgical resection12,13 and temozolomide (TMZ).14,15 It is estimated that <10% of glioblastomas harbor the IDH

EP

mutations16 while 30-68% of AAs harbor IDH mutations.17,13,18,19 The assumption that therapeutic interventions would produce similar results in glioblastoma and AA, thus, warrant

AC C

careful scrutiny.

While there is a large body literature providing class II/III data demonstrating the clinical benefit of gross total resection (GTR) for glioblastomas, the literature supporting such benefit for AAs is more limited.20,21 A previous study using the SEER database demonstrated a reduced hazard ratio for death in AA patients who underwent GTR relative to those who underwent surgical

ACCEPTED MANUSCRIPT

Page 4 biopsy.22 However, the efficacy of GTR relative to STR in this population was not carefully studied.23 Moreover, whether these effects differ between glioblastoma and AA patients remains

RI PT

unknown. It is generally accepted that AA, like glioblastoma, is an intrinsically infiltrative

disease and that microscopic total resection is not possible without significant morbidity.24

However, whether AAs and glioblastomas differ in their infiltrative nature or differ in their

SC

response to TMZ therapy remains an open question.25,26 If such differences exist, one may expect that the benefit of GTR in the AA population may differ from that seen in the glioblastoma

M AN U

population.

In this context, we explored whether the clinical benefit of surgical resection in AA patients differed from those of glioblastoma patients. Because of the rarity of AAs, we utilized the Surveillance, Epidemiology, and End Results (SEER) population database to identify a cohort of

TE D

2,755 AA patients and 21,962 glioblastoma patients. We found that the AA patients derive notably greater benefit from gross total resection (GTR) relative to glioblastoma patients, in both relative terms as assessed by hazard of death and in absolute terms as gauged by median

AC C

patients.

EP

survival. This finding bears relevance to surgical decision-making during treatment of AA

MATERIALS AND METHODS Data and Study Population

ACCEPTED MANUSCRIPT

Page 5 The Surveillance, Epidemiology, and End Results (SEER) Program was established by the National Cancer Institute (NCI) to collect cancer incidence and survival data from 18 population-

RI PT

based cancer registries that cover approximately 28% of total U.S. population (SEER Research Data 1973-2010). We downloaded the dataset as ASCII text files released in April 2013 based on

SC

the November 2012 submission.27

This study included patients who were diagnosed between 1999 and 2010 with WHO grade III-

M AN U

IV intracranial astrocytomas as the only cancer diagnosis. We used International Classification of Disease for Oncology-third edition (ICD-O-3) histology codes 9401 for WHO Grade III anaplastic astrocytoma and 9440-9442 for WHO Grade IV glioblastoma. ICD-O-3 topologic site codes C71.0-C71.9 were used to select for brain tumors. These codes were cross-referenced and validated with Table 1 of Central Brain Tumor Registry of the United States (CBTRUS)

TE D

Statistical Report.1 Patients were excluded from the study if the surgical status was coded as unknown or if the histology was coded as unconfirmed. After these exclusions, we identified a

EP

total of 2,755 AA and 21,962 glioblastoma cases.

AC C

Covariates and Extent of Resection

Survival time was defined as the number of months from diagnosis to the date of death due to any cause or the date of last known follow-up. We used the following demographic variables in the statistical analysis: age (<18, 18-44, 45-49, 50-54, 55-59, 60-74, or >75 years), race/ethnicity (White, Black, Asian/Pacific Islander, Hispanic, American Indian/Alaskan Native, or

ACCEPTED MANUSCRIPT

Page 6 Other/Unknown), marital status (single, married, or [separated, divorced or widowed]), and sex (male or female). Clinical variables included tumor size (<5cm, 5-7cm or >7cm), tumor location

RI PT

(based on ICD-O-3 topologic site codes C71.0-C71.9), radiotherapy status (treatment or no treatment) and surgical treatment received (no surgery, subtotal resection or gross total

SC

resection).

With regards to the extent of resection achieved, we utilized the following surgery codes from

M AN U

the SEER registry: no surgery (code 00), excisional biopsy (code 20), subtotal or partial resection (codes 21, 40), or gross total resection (GTR) (code 30, 55). While the exact definition of surgical codes underwent minor modifications with each edition of SEER Program Coding and Staging Manual (1998-2003, 2004-2006, 2007-2009, 2010-present), the general definition remained consistent throughout the various editions.28,29 The latest definition for surgical codes

TE D

used at the time of this study can be found in the SEER Program Coding and Staging Manual 2013 released on February 28, 2013 under Appendix C: Surgical Codes for Brain.30 Historical definitions can also be found on the SEER website. 31 GTR, STR, and biopsy were defined as

EP

derived in previously published manuscripts that examined the relative efficacy of GTR and STR

AC C

in SEER glioblastoma.28,29

Statistical Analysis

All analyses were conducted using Stata version 11.2,32 and the level of statistical significance was set at p < 0.05. We performed comparisons of overall survival between AA and

ACCEPTED MANUSCRIPT

Page 7 glioblastoma patients who underwent STR, GTR, or no surgery. Further analyses were performed to determine the impact of age and TMZ use on the survival benefit associated with

RI PT

GTR. We used the Kaplan-Meier method to generate unadjusted survival curves in both overall and subset analyses. Statistical significance was determined using log-rank test across survival functions.28 To obtain the multivariate adjusted hazard ratio (HR) of death, we performed Cox

SC

proportional hazard analysis adjusting for demographic and clinical covariates mentioned above.

overall and subset analyses.

RESULTS

TE D

Patient and Clinical Characteristics

M AN U

Additionally, we calculated the median survival with 95% confidence interval (CI) in both

Patient characteristics are outlined in Table 1. From the SEER database, we identified 21,962 glioblastoma patients and 2,775 AA patients. The median age of diagnosis (interquartile

EP

range) was 50 (35-64) for anaplastic astrocytomas and 61 (52-71) for glioblastomas. 9,057

AC C

(41.24%) of anaplastic astrocytoma and 13,272 (42.37%) of glioblastoma patients were female. Mortality rates increased with increasing histologic grade with 67.4% mortality for anaplastic astrocytoma and 85.85% for glioblastoma. The most common sites for both anaplastic astrocytoma and glioblastoma were the frontal lobe and the temporal lobe. The demographics of this population are consistent with previous literature.2

ACCEPTED MANUSCRIPT

Page 8 Of the 2,775 patients with anaplastic astrocytoma, 1,091 (39.6%) had no surgery, 503 had biopsy (18.26%), 624 (22.65%) had STR, and 537 (19.49%) had GTR. Of the 9,057 patients with

RI PT

glioblastoma, 4,718 (21.48%) had no surgery, 4,459 (20.3%) had biopsy, 6,170 (28.09%) had

SC

STR, and 6,615 (30.12%) had GTR.

Multivariate Adjusted HR of Death Analysis by Extent of Resection

M AN U

We analyzed survival trends for patients who had no surgery, STR or GTR (Figure 1). We derived adjusted hazard ratios (HRs) for death using an extended multivariate Cox proportional hazards model to determine whether the improvements in survival persisted after adjusting for age, race/ethnicity, marital status, sex, year of diagnosis, tumor size, tumor site, and

TE D

radiotherapy (Table 2). The risk of death as assessed by Cox proportional hazard ratio for AA patients who underwent GTR was 0.599 while those who underwent STR was 1. The difference between these hazard ratios was significant at p<0.0001. The hazard ratio of death for

EP

glioblastoma patients who underwent GTR was 0.762 compared to 1.00 for glioblastoma patients who underwent STR (p<0.0001). We further compared the median survival in months between

AC C

the two groups. The median survival for AA patients who underwent GTR was 64 months compared to the median survival of AA patients who underwent STR was 24 months (p<0.0001, Figure 2). The median survival for glioblastoma patients who underwent GTR was 13 months compared to the median survival of glioblastoma patients who underwent STR was 9 months (p<0.0001).

ACCEPTED MANUSCRIPT

Page 9 Multivariate adjusted HR of Death by Extent of Resection and Age

RI PT

To understand the effect of age on the survival benefit of GTR, we divided the SEER patients into patients <50 years old and patients ≥50 years old based on age cut-offs on the established recursive partitioning classification (RPA) classification scheme.33 We saw a

SC

baseline increase in the HR of death in the age >50 patients for both AAs and glioblastomas in all categories (Table 3). The survival benefit of GTR was greatest in AA patients age <50 who

M AN U

underwent GTR relative to those who underwent STR. In this population, the HR for death was reduced by 66% for GTR patients (HR=0.445 [0.326-0.608]) relative to those undergoing STR. The median survival of AA patients age <50 who underwent GTR and STR were not reached and 42 months respectively. For glioblastoma, the HR for death for patients age <50 was reduced only by 20% (H.R=0.824 [0.742-0.914]) relative to those undergoing STR. The median

respectively (Figure 3).

TE D

survival of glioblastoma patients age <50 who underwent GTR and STR were 19 and 15 months

EP

Multivariate adjusted HR of Death by Extent of Resection and Temozolomide (TMZ) Use To understand how use of Temozolomide (TMZ) influenced the survival association

AC C

between GTR and overall survival, we analyzed our results as a function of time based on the introduction of TMZ in 2005.7 We determined the survival benefit of GTR relative to STR in the pre-TMZ era (1999-2004) and the post-TMZ (2005-2010). The overall HR of death for both AA and glioblastoma patients was lower in the post-TMZ era irrespective of the extent of resection, corroborating the published efficacy of TMZ treatment for both AAs and glioblastomas (Table

ACCEPTED MANUSCRIPT

Page 10 4). The survival benefit of GTR was greatest in AA patients who underwent GTR in the postTMZ era. In this population, the HR for death was reduced by two-fold (HR=0.367 [0.278-

RI PT

0.502]) relative to those undergoing STR. The median survival of AA patients who underwent GTR and STR in the post-TMZ era was not reached for GTR patients and 33 months for STR patients, respectively. For glioblastoma patients in the post-TMZ era, the HR for death was

SC

reduced only by 22% (HR=0.582 [0.546-0.620]) relative to those undergoing STR. The median survival of glioblastoma patients who underwent GTR and STR in the post-TMZ era were 14

M AN U

and 10 months respectively (Figures 4a & 4b).

DISCUSSION

TE D

To our knowledge, our study represents the first effort to quantitatively compare the survival benefit of GTR relative to STR for AA and glioblastoma patients using a large population database. While GTR is associated with improved benefit in both AA and glioblastoma relative

EP

to STR, the magnitude of benefit was notably greater in AA patients. The net reduction in the hazard ratio (HR) of death for AA patients undergoing GTR relative to STR was 59% greater

AC C

than the ratio observed for glioblastoma patients (p<0.001) suggesting AA patients would benefit more from GTR than glioblastoma patients. The difference in median survival for AA patients based on EOR was >3 years, as AA patients who underwent STR exhibited a median survival of 2 years whereas the median survival for AAs who underwent GTR was > 5 years. In comparison,

ACCEPTED MANUSCRIPT

Page 11 there was only an average difference of 4 months between glioblastoma patients who underwent

RI PT

GTR and STR. These observations should impact clinical decision making during the planning of surgical resection. A survival expectation of three months versus several years for AA versus

glioblastoma GTR may significantly impact the surgeon’s risk benefit analysis, particularly if

SC

GTR is likely to incur supplemental motor syndrome34 or other deficits that require lengthy

M AN U

recovery.35 In this context, the development of a rapid intraoperative test to determine a tumor’s histopathology would be helpful in clinical decision making.

Our analyses suggest that the benefit of GTR in AA patients is most notable for patients younger than 50. The age-survival association likely reflects the observation that patients with IDH

TE D

mutated tumors derived greater benefit from GTR relative to IDH wild type tumors.13 Consistent with this clinical observation, IDH mutated tumors have been shown to be less infiltrative relative to the IDH wild type tumors.36 GTR of IDH mutated tumors should, thus, likely leave

EP

lower residual tumor burden relative to GTR of IDH wild type tumors. IDH mutated gliomas tend to occur in patients <50 years old.18,37 In this context, our data suggest that AA patients age

AC C

<50 may be a proxy for IDH1 mutation. While this hypothesis may be tested by examining the survival of GTR in this population after controlling for IDH mutation status, such analysis is not feasible given the absence of molecular information in the SEER database. It is important to note that there has not been a dedicated randomized controlled study characterizing the effectiveness of TMZ in AA patients. In the landmark Stupp study,38 AA

ACCEPTED MANUSCRIPT

Page 12 patients were combined with glioblastoma patients – with efficacy demonstrated in AA patients in a post-hoc analysis. The improved survival in the post-TMZ era largely supported the

RI PT

effectiveness of TMZ in AA patients. After GTR, a major determinant of patient survival is whether the residual microscopic cancer cells respond to the tumor ablative effects of

chemotherapy and radiation therapy. Since the use of radiation therapy did not significantly

SC

differ in the pre- and post-TMZ era,29 our results suggest that patients treated in the TMZ era derived survival benefit relative to the pre-TMZ era and that this survival benefit is likely due to

M AN U

the effectiveness of TMZ against the microscopic cancer cells that remain after GTR. A major strength of this report involves the number of AAs studied. As a rare cancer, it is difficult to recapitulate the number of AAs studied here through institutional experiences or clinical trials.39 Nevertheless, there are several limitations to this study that warrant serious

TE D

consideration. First, there is likely a selection bias in those patients who underwent GTR relative to STR. Second, the study is a retrospective study of a population-based cancer registry and is subject to the various flaws inherent to such a study. Third, the SEER dataset does not harbor

EP

case volume information for the various participating centers. As such, we cannot examine the

AC C

impact of clinical volume on the observed effect. Finally, the SEER dataset is limited by the absence of key clinical variables such as patient chemotherapeutic regimens. However, we do not believe that the absence of such variables confound the results reported here since nearly all AA patients undergo concurrent chemo/radiation therapy irrespective of whether they received GTR or STR. As such, we feel that our conclusion is robust.

ACCEPTED MANUSCRIPT

Page 13 The high proportion of AA patients who did not undergo surgical biopsy in the SEER database is alarming, particularly in the context of the modern oncology paradigm grounded on microscopic

RI PT

and molecular scrutiny of pathologic tissue samples.40 To the extent that the anatomic

localization of AAs did not significantly differ from that of glioblastoma, we feel that this high proportion is unlikely related to ease or safety of surgical access. Moreover, given the safety of

SC

stereotactic biopsy,41 there are few clinical scenarios in which treatment based on presumptive

standard of surgical care for AA patients.

M AN U

diagnosis is warranted. Further studies of this matter are warranted in order to optimize the

Our study suggests the critical need for two areas of development in surgical neuro-oncology. In the context of this study, we recognize that incorporating genomic and quality of life data into large population databases such as SEER are critical next steps to studying the impact of clinical

TE D

practice (e.g. surgical resection). Second, technologies are needed to afford neurosurgeons access to definitive histologic diagnosis in real-time and opportunities to refine surgical plan

AC C

CONCLUSION

EP

based on this information.

Within the limitations of a cancer registry based study, we demonstrate that AA patients who underwent GTR derived survival benefit relative to those in whom GTR was not achieved or attempted. The magnitude of this survival benefit was especially notable in AA patients relative

ACCEPTED MANUSCRIPT

Page 14 to that observed in glioblastoma patients. Clinical decision making regarding the extent of

AC C

EP

TE D

M AN U

SC

RI PT

resection should take these observations into consideration.

ACCEPTED MANUSCRIPT

Page 15 References Ostrom QT, Gittleman H, Liao P, et al. CBTRUS Statistical Report: Primary Brain and Central Nervous System Tumors Diagnosed in the United States in 2007–2011. Neuro Oncol. 2014;16(suppl 4):iv1-iv63.

2.

Ohgaki H, Kleihues P. Population-based studies on incidence, survival rates, and genetic alterations in astrocytic and oligodendroglial gliomas. J Neuropathol Exp Neurol. 2005;64(6):479-489.

3.

Johnson DR, O’Neill BP. Glioblastoma survival in the United States before and during the temozolomide era. J Neurooncol. 2012;107(2):359-364.

4.

Kramm CM, Wagner S, Van Gool S, et al. Improved survival after gross total resection of malignant gliomas in pediatric patients from the HIT-GBM studies. Anticancer Res. 2006;26(5B):3773-3779.

5.

Linda M Mcmanus Richard, ed. Pathobiology of Human Disease: A Dynamic Encyclopedia of Disease Mechanisms. Elsevier; 2014:3592-3606.

6.

Ahmed Rasheed BK, Wiltshire RN, Bigner SH, Bigner DD. Molecular pathogenesis of malignant gliomas. Curr Opin Oncol. 1999;11(3):162.

7.

Larner J, Jane J, Laws E, Packer R, Myers C, Shaffrey M. A phase I-II trial of lovastatin for anaplastic astrocytoma and glioblastoma multiforme. Am J Clin Oncol. 1998;21(6):579-583.

8.

Gorovets D, Kannan K, Shen R, et al. IDH mutation and neuroglial developmental features define clinically distinct subclasses of lower grade diffuse astrocytic glioma. Clin Cancer Res. 2012;18(9):2490-2501.

9.

Turkalp Z, Karamchandani J, Das S. IDH mutation in glioma: new insights and promises for the future. JAMA Neurol. 2014;71(10):1319-1325.

AC C

EP

TE D

M AN U

SC

RI PT

1.

10. Dunn GP, Rinne ML, Wykosky J, et al. Emerging insights into the molecular and cellular basis of glioblastoma. Genes Dev. 2012;26(8):756-784. 11. Losman J-A, Kaelin WG Jr. What a difference a hydroxyl makes: mutant IDH, (R)-2hydroxyglutarate, and cancer. Genes Dev. 2013;27(8):836-852. 12. Juratli TA, Cahill DP, McCutcheon IE. Determining optimal treatment strategy for diffuse glioma: the emerging role of IDH mutations. Expert Rev Anticancer Ther. 2015;15(6):603606.

ACCEPTED MANUSCRIPT

Page 16 13. Beiko J, Suki D, Hess KR, et al. IDH1 mutant malignant astrocytomas are more amenable to surgical resection and have a survival benefit associated with maximal surgical resection. Neuro Oncol. 2014;16(1):81-91.

RI PT

14. Yoshimura J, Onda K, Tanaka R, Takahashi H. Clinicopathological study of diffuse type brainstem gliomas: analysis of 40 autopsy cases. Neurol Med Chir . 2003;43(8):375-382; discussion 382. 15. Houillier C, Wang X, Kaloshi G, et al. IDH1 or IDH2 mutations predict longer survival and response to temozolomide in low-grade gliomas. Neurology. 2010;75(17):1560-1566.

SC

16. Yan H, Parsons DW, Jin G, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009;360(8):765-773.

M AN U

17. Balss J, Meyer J, Mueller W, Korshunov A, Hartmann C, von Deimling A. Analysis of the IDH1 codon 132 mutation in brain tumors. Acta Neuropathol. 2008;116(6):597-602. 18. Hartmann C, Meyer J, Balss J, et al. Type and frequency of IDH1 and IDH2 mutations are related to astrocytic and oligodendroglial differentiation and age: a study of 1,010 diffuse gliomas. Acta Neuropathol. 2009;118(4):469-474.

TE D

19. Mukasa A, Takayanagi S, Saito K, et al. Significance of IDH mutations varies with tumor histology, grade, and genetics in Japanese glioma patients. Cancer Sci. 2012;103(3):587592. 20. Sanai N, Berger MS. Glioma extent of resection and its impact on patient outcome. Neurosurgery. 2008;62(4):753-764; discussion 264-266.

EP

21. Michel Lacroix, Dima Abi-Said, Daryl R. Fourney, et al. A multivariate analysis of 416 patients with glioblastoma multiforme: prognosis, extent of resection, and survival. J Neurosurg. 2001;95(2):190-198.

AC C

22. Nuño M, Drazin DG, Acosta FL Jr. Differences in treatments and outcomes for idiopathic scoliosis patients treated in the United States from 1998 to 2007: impact of socioeconomic variables and ethnicity. Spine J. 2013;13(2):116-123. 23. Sanai N. Emerging operative strategies in neurosurgical oncology. Curr Opin Neurol. 2012;25(6):756-766. 24. McCrea HJ, Bander ED, Venn RA, et al. Sex, Age, Anatomic Location, and Extent of Resection Influence Outcomes in Children With High-grade Glioma. Neurosurgery. 2015;77(3):443-453. 25. SongTao Q, Lei Y, Si G, et al. IDH mutations predict longer survival and response to

ACCEPTED MANUSCRIPT

Page 17 temozolomide in secondary glioblastoma. Cancer Sci. 2012;103(2):269-273.

RI PT

26. Chinot OL, Honore S, Dufour H, et al. Safety and efficacy of temozolomide in patients with recurrent anaplastic oligodendrogliomas after standard radiotherapy and chemotherapy. J Clin Oncol. 2001;19(9):2449-2455. 27. SEER Data 1973-2010.

28. Noorbakhsh A, Tang JA, Marcus LP, et al. Gross-total resection outcomes in an elderly population with glioblastoma: a SEER-based analysis. J Neurosurg. 2014;120(1):31-39.

M AN U

30. SEER Appendix C: Site Specific Coding Modules. http://seer.cancer.gov/manuals/2013/appendixc.html.

SC

29. Dong X, Noorbakhsh A, Hirshman BR, et al. Survival trends of grade I, II, and III astrocytoma patients and associated clinical practice patterns between 1999 and 2010: A SEER-based analysis. Neuro-Oncology Practice. July 2015. doi:10.1093/nop/npv016.

31. SEER Historical Staging and Coding Manuals. http://seer.cancer.gov/tools/codingmanuals/historical.html.

32. StataCorp. 2009. Stata Statistical Software: Release 11. College Station, TX: StataCorp LP.

TE D

33. Curran WJ Jr, Scott CB, Horton J, et al. Recursive partitioning analysis of prognostic factors in three Radiation Therapy Oncology Group malignant glioma trials. J Natl Cancer Inst. 1993;85(9):704-710.

EP

34. Russell SM, Kelly PJ. Incidence and clinical evolution of postoperative deficits after volumetric stereotactic resection of glial neoplasms involving the supplementary motor area. Neurosurgery. 2003;52(3):506-516; discussiom 515-516.

AC C

35. Chang SM, Parney IF, Huang W, et al. Patterns of care for adults with newly diagnosed malignant glioma. JAMA. 2005;293(5):557-564. 36. Nobusawa S, Watanabe T, Kleihues P, Ohgaki H. IDH1 mutations as molecular signature and predictive factor of secondary glioblastomas. Clin Cancer Res. 2009;15(19):6002-6007. 37. Kim W, Liau LM. IDH mutations in human glioma. Neurosurg Clin N Am. 2012;23(3):471480. 38. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352(10):987-996. 39. Stupp R, Reni M, Gatta G, Mazza E, Vecht C. Anaplastic astrocytoma in adults. Crit Rev

ACCEPTED MANUSCRIPT

Page 18 Oncol Hematol. 2007;63(1):72-80.

RI PT

40. Huse JT, Holland EC. Targeting brain cancer: advances in the molecular pathology of malignant glioma and medulloblastoma. Nat Rev Cancer. 2010;10(5):319-331.

AC C

EP

TE D

M AN U

SC

41. Waters JD, Gonda DD, Reddy H, Kasper EM, Warnke PC, Chen CC. Diagnostic yield of stereotactic needle-biopsies of sub-cubic centimeter intracranial lesions. Surg Neurol Int. 2013;4(Suppl 3):S176-S181.

ACCEPTED MANUSCRIPT

Page 19

RI PT

Figure Legend:

Table 1. Demographic and clinical characteristic of AA and glioblastoma cases, SEER

SC

1999-2010

M AN U

Figure 1a. Extent of resection (EOR) achieved for AA patients

TE D

Figure 1b. Extent of resection (EOR) achieved for glioblastoma patients

EP

Table 2. Multivariate adjusted HR* of death by EOR in AA and glioblastoma patients

AC C

Figure 2. Kaplan-Meier plot of 5-year survival by AA and glioblastoma patients

Table 3. Multivariate adjusted HR* of death by EOR in AA and glioblastoma for patients < or > 50 years of age

ACCEPTED MANUSCRIPT

Page 20

RI PT

Figure 3a. Kaplan-Meier plot of 5-year survival by EOR in AA and glioblastoma for patients < 50 years of age

SC

Figure 3b. Kaplan-Meier plot of 5-year survival by EOR in AA and glioblastoma for

M AN U

patients > 50 years of age

Table 4. Multivariate adjusted HR* of death by EOR in AA and glioblastoma for patients

TE D

in pre-TMZ and post-TMZ era

Figure 4a. Figure 4. Kaplan-Meier plot of 5-year survival by EOR in AA and glioblastoma

EP

for patients in pre-TMZ era

AC C

Figure 4b. Kaplan-Meier plot of 5-year survival by EOR in AA and glioblastoma for patients in post-TMZ era

ACCEPTED MANUSCRIPT

Table 1. Demographic and clinical characteristic of AA and glioblastoma cases, SEER 1999-2010

AC C

EP

RI PT

Glioblastoma 21,962 (70.11) 61 (52-71) n = 21,962 6,615 (30.12) 6,170 (28.09) 4,459 (20.3) 4,718 (21.48) n = 21,483 5,353 (24.92) 16,130 (75.08) n = 21,962 255 (1.16) 2,256 (10.27) 1,757 (8.00) 2,460 (11.20) 2,962 (13.49) 8,520 (38.79) 3,752 (17.08) n = 21,962 17,665 (80.43) 1,213 (5.52) 845 (3.85) 2,137 (9.73) 66 (0.3) 36 (0.16) n = 21,311 3,008 (14.11) 14,165 (66.47) 4,138 (19.42) n = 21,962 12,905 (58.76) 9,057 (41.24) n = 16,847 9,630 (57.16) 5,728 (34) 1,489 (8.84) n = 21,962 5,840 (26.59)

SC

M AN U

TE D

Number of Patients, No. (% of Total) Age, median (IQR) Surgery, No. (%) Gross Total Resection Partial Resection Local Excision/Biopsy No Surgery Radiotherapy, No. (%) No Yes Age Category, No. (%) Age < 18 Age 18-44 Age 45-49 Age 50-54 Age 55-59 Age 60-74 Age ≥ 75 Race, No. (%) White Black Asian/Pacific Islander Hispanic American Indian/Alaskan Native Other/Unknown, Non-Hispanic Marital Status, No. (%) Single Married Separated, Divorced, Widowed Sex, No. (%) Male Female Tumor Size, No. (%) <5cm 5-7cm >7cm Tumor Site, No. (%) Frontal Lobe

Anaplastic Astrocytoma 2,755 (29.89) 50 (35-64) n = 2,775 537 (19.49) 624 (22.65) 503 (18.26) 1,091 (39.6) n = 2,698 611 (22.65) 2,087 (77.35) n = 2,775 177 (6.42) 929 (33.72) 249 (9.04) 231 (8.38) 265 (9.62) 593 (21.52) 311 (11.29) n = 2,775 2,097 (76.12) 180 (6.53) 138 (5.01) 325 (11.8) 7 (0.25) 8 (0.29) n = 2,669 697 (26.11) 1,581 (59.24) 391 (14.65) n = 2,775 1,533 (55.64) 1,222 (44.36) n = 1,744 1,107 (63.47) 429 (24.6) 208 (11.93) n = 2,775 861 (31.25)

ACCEPTED MANUSCRIPT

RI PT

5,377 (24.48) 3,691 (16.81) 952 (4.33) 133 (0.61) 3713 (16.91) 851 (3.87) 1,154 (5.25) 95 (0.43) 156 (0.71) n = 21,962 873 (3.98) 1728 (7.87) 1708 (7.78) 1776 (8.09) 1861 (8.47) 1967 (8.96) 1934 (8.81) 1877 (8.55) 2026 (9.23) 2063 (9.39) 2041 (9.29) 2108 (9.6) n = 21,962 3,108 (14.15) 18,854 (85.85)

SC

531 (19.27) 323 (11.72) 61 (2.21) 88 (3.19) 405 (14.7) 272 (9.87) 136 (4.94) 32 (1.16) 46 (1.67) n = 2,775 103 (3.74) 254 (9.22) 227 (8.24) 240 (8.71) 250 (9.07) 241 (8.75) 246 (8.93) 242 (8.78) 225 (8.17) 229 (8.31) 245 (8.89) 253 (9.18) n = 2,775 898 (32.6) 1,857 (67.4)

M AN U

TE D

Temporal Lobe Parietal Lobe Occipital Lobe Brain Stem Overlapping Lesion of Brain Cerebrum Brain, NOS Ventricle, NOS Cerebellum, NOS Year of Diagnosis, No. (%) 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Overall Mortality, No. (%) Living Deceased

AC C

EP

Abbreviations: Abbreviations: SEER, Surveillance, Epidemiology, and End Results; IQR, Interquartile range; NOS, Not otherwise specified.

ACCEPTED MANUSCRIPT

Table 2. Multivariate adjusted HR* of death by EOR in AA and glioblastoma patients

Adjusted HR 1.00 0.599 (0.488 - 0.735) 1.488 (1.256 - 1.762)

P value Reference <0.0001 <0.0001

Glioblastoma

RI PT

Anaplastic Astrocytoma Overall HR (19992010) Local excision/STR GTR No Surgery

Adjusted HR 1.00 0.762 (0.729 - 0.798) 1.552 (1.471 – 1.637)

P value Reference <0.0001 <0.0001

M AN U

SC

*Adjusted for age, race/ethnicity, marital status, sex, year of diagnosis, tumor size, tumor site, and radiotherapy.

Table 3. Multivariate adjusted HR* of death by EOR in AA and glioblastoma for patients < or > 50 years of age

TE D

Age < 50 Local excision/STR GTR No Surgery Age > 50 Local excision/STR GTR No Surgery

Anaplastic Astrocytoma Adjusted HR P value 1.00 Reference 0.445 (0.326 - 0.608) <0.0001 1.634 (1.259 - 2.120) <0.0001 Adjusted HR P value 2.671 (2.041 - 3.495) <0.0001 2.495 (1.872 - 3.325) <0.0001 4.431 (3.509 - 5.595) <0.0001

Glioblastoma Adjusted HR P value 1.00 Reference 0.824 (0.742 - 0.914) <0.0001 1.598 (1.389 - 1.840) <0.0001 Adjusted HR P value 2.086 (1.911 - 2.277) <0.0001 1.549 (1.419 - 1.691) <0.0001 3.329 (3.038 - 3.648) <0.0001

EP

*Adjusted for race/ethnicity, marital status, sex, year of diagnosis, tumor size, tumor site, and radiotherapy.

AC C

Table 4. Multivariate adjusted HR* of death by EOR in AA and glioblastoma for patients in pre-TMZ and post-TMZ era Pre-TMZ Local excision/STR GTR No Surgery Post-TMZ Local excision/STR GTR No Surgery

Anaplastic Astrocytoma Glioblastoma Adjusted HR P value Adjusted HR P value 1.00 Reference 1.00 Reference 0.615 (0.477 - 0.792) <0.0001 0.741 (0.695 - 0.791) <0.0001 1.369 (1.101 - 1.704) <0.0001 1.449 (1.343 - 1.564) <0.0001 Adjusted HR P value Adjusted HR P value 0.650 (0.498 - 0.848) <0.0001 0.749 (0.701 - 0.799) <0.0001 0.367 (0.268 - 0.502) <0.0001 0.582 (0.546 - 0.620) <0.0001 1.064 (0.855 - 1.324) 0.258 1.237 (1.150 - 1.330) <0.0001

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

*Adjusted for age, race/ethnicity, marital status, sex, tumor size, tumor site, and radiotherapy.

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

Highlights:

AC C

EP

TE D

M AN U

SC

RI PT

1. This is the first study documenting the clinical benefit of GTR specifically for anaplastic astrocytomas compared to glioblastomas. 2. AA patients who underwent GTR had an increased survival rate compared to AA patients who underwent STR. 3. The survival benefit associated with GTR was greater for AA patients relative to glioblastoma patients 4. The survival benefit of GTR in AA was most pronounced for patients <50 years old.

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

SEER: Surveillance, Epidemiology, and End Results IQR: Interquartile range NOS: Not otherwise specified. AA: Anaplastic Astrocytoma GTR: Gross total resection STR: Subtotal resection IDH1: Isocitrate dehydrogenase HR: Hazard Ratio

RI PT

Abbreviations: