Effect of Treatment Modalities on Progression-Free Survival and Overall Survival in Molecularly Subtyped World Health Organization Grade II Diffuse Gliomas: A Systematic Review

Journal Pre-proof Effect of Treatment Modalities on Progression-Free Survival and Overall Survival, in Molecularly Subtyped WHO Grade II Diffuse Gliomas: A Systematic Review Arash Ghaffari-Rafi, B.S., George Samandouras, M.D. PII:

S1878-8750(19)32267-3

DOI:

https://doi.org/10.1016/j.wneu.2019.08.111

Reference:

WNEU 13147

To appear in:

World Neurosurgery

Received Date: 2 June 2019 Revised Date:

16 August 2019

Accepted Date: 17 August 2019

Please cite this article as: Ghaffari-Rafi A, Samandouras G, Effect of Treatment Modalities on Progression-Free Survival and Overall Survival, in Molecularly Subtyped WHO Grade II Diffuse Gliomas: A Systematic Review, World Neurosurgery (2019), doi: https://doi.org/10.1016/j.wneu.2019.08.111. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. 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. © 2019 Elsevier Inc. All rights reserved.

1

Title Page Effect of Treatment Modalities on Progression-Free Survival and Overall Survival, in Molecularly Subtyped WHO Grade II Diffuse Gliomas: A Systematic Review Arash Ghaffari − Rafi , B.S., George Samandouras , M.D. 1. University of Hawai’i at Mānoa, John A. Burns School of Medicine (Honolulu, Hawaii USA) 2. University College London, Queen Square Institute of Neurology (London, England UK) 3. University College London Hospitals NHS Trust, Victor Horsley Department of Neurosurgery, The National Hospital for Neurology and Neurosurgery (London, England UK)

# Author for correspondence: Arash Ghaffari-Rafi Email: [email protected] Present Postal Address: 465 Caledonian Road Room BL08 London, United Kingdom N7 9BA Keywords: low grade glioma; astrocytoma; oligodendroglioma; glioma; progression free survival; overall survival Word count (body): 3898

2

Abstract Objective With the 2016 update of the WHO Classification of Tumors of the Central Nervous System incorporating molecular subtyping to histology, WHO grade II diffuse astrocytic and oligodendroglial tumors are now subcategorized by distinct molecular markers. Currently, there are no published systematic reviews quantifying differences in progression free survival (PFS) and overall survival (OS) on the basis of molecular subtypes of WHO grade II diffuse gliomas, against the background of administered treatments. Methods Utilizing the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) and the Cochrane Handbook of Systemic Reviews of Interventions, we conducted a systematic review through MEDLINE, Embase, and CENTRAL. Results For OS, the first quartile (25%), median (50%), third quartile (75%), and 95% confidence interval were respectively identified (in months)—astrocytoma-wild type WHO II (A-wt II): 22.8, 32.2, 40.7, and 21.6-61.2; astrocytoma-mutant WHO II (A-mt II): 69.85, 115.2, 128.4, and 55.4-164.0; oligodendroglioma WHO II (OD II): 106.3, 163.7, 213.3, and 67.3-235.4 (p-value = 0.0001675). For PFS, the 25th, 50th, and 75th percentiles, and 95% confidence interval are as follows (in months), respectively—A-wt II: 6.90, 17.45, 19.57, and 3.00-23.69; A-mt II: 37.20, 43.20, 55.63, and 35.7-60.0; OD II: 47.42, 59.2, 88.28, and 46.3-91.2 (p-value = 0.01488). Conclusion This appears to be the first systematic review of OS and PFS in patients with WHO grade II LGGs, against treatment modalities, in molecularly stratified subsets introduced by the WHO 2016 classification of CNS tumors. Overall, A-wt II was confirmed to have a significantly shorter OS than A-mt II; meanwhile, there was no significant difference found between OS of OD II with Awt II and A-mt II. Additionally, all three molecular subtypes were found to have statistically significant differences between PFS, with OD II having a statistically better PFS than A-mt II. These data can provide valuable prognostic insight to patients and clinicians. Additionally, assessing survival differences enhances understanding of treatment recommendations against molecular markers and may facilitate future clinical trial design.

3

1.1 Introduction Aside generalizations, accurate epidemiological data on World Health Organization (WHO) grade II astrocytic and oligodendroglial tumours (low grade gliomas [LGGs]) remains elusive. Studies estimate annually about 2,000 to 3,000 LGGs are diagnosed in the United States [1, 2]. Affecting mostly fully functioning patients, usually in the second to fourth decades of life, combined with a natural history composed of clinical and radiological progression with an unpredictable malignant transformation, the actual burden of LGGs is considerable [3, 4]. In addition, due to lack of randomized controlled trials (RTCs) comparing multiple treatment modalities, their optimal management remains disputed, ranging from serial imaging to attempts of maximum safe resection, with combinations of neo-adjuvant or post-operative chemoradiotherapy [1]. More recently, compared to the 2007 predecessor, the 2016 update of the WHO classification of Central Nervous System (CNS) tumors, has incorporated molecular features with histology to create an integrated diagnosis [5]. Current diagnostic standards require determining status of isocitrate dehydrogenase (IDH) mutations and codeletion in chromosomal arms 1p and 19q [5]. Moreover, data supports the necessity to stratify survival outcomes by molecular subtypes—for one phase III trial demonstrated similar survival between initial treatment with chemotherapy (CT) versus radiation therapy (RT), until outcomes were analyzed with molecular diagnostics [6]. Since the 2016 WHO classification of LGGs, the current study appears to be the first comprehensive systematic review quantifying differences in progression free survival (PFS) and overall survival (OS) of molecularly stratified WHO grade II diffuse gliomas. By assessing differences in survival associated with surgery, radiation therapy, and chemotherapy, a better understanding can be developed on how molecular features influence treatment recommendations and can subsequently facilitate future clinical trial design.

2.1 Materials and Methods This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) and the Cochrane Handbook of Systematic Reviews of Interventions [7, 8, 9].

2.1.1 Eligibility Criteria Study Types Only clinical studies were included, with experimental and animal studies excluded. Participants Studies including data stratified for adult humans (18 years or older) with molecularly subtyped newly diagnosed WHO grade II diffuse gliomas (astrocytoma or oligodendroglioma) were included [5]. The genotypes were defined as follows: wild type astrocytomas, IDH-wild type;

4 mutant astrocytoma, IDH-mutant with non-1p/19q codeletion; oligodendroglioma, IDH-mutant with 1p/19q codeletion. Moreover, for studies that provided grade II gliomas with only positive codeletion (without IDH status), such tumors were assumed to be oligodendrogliomas, on the basis of most 1p/19q codeletion patients also having IDH-mt [10]. Interventions Interventions targeting WHO grade II diffuse gliomas were not limited to any subcategory, and included surgery, chemotherapy, and radiotherapy. Outcomes The endpoints of overall survival and progression-free survival, in the unit of time (days, months, years) or rate, were collected. Our study defined OS as the time of intervention to death by any cause. PFS was defined as the time of intervention to tumor recurrence/progression, evidenced by radiological (magnetic resonance imaging [MRI]) or clinical deterioration. Clinical deterioration would include development of new or worsening of existing focal deficits or symptoms of elevated intracranial pressure. Radiologic deterioration would include increased or new tumor contrast enhancement, enlargement of volume, increasing mass effect, or increasing midline shift. Due to possibility of variation in outcome measure definitions, from each study the outcome measure definition was also collected. Follow Up Time Follow-up time was restricted to 48 months and collected. Language Only articles written in English were included. Information Sources Medical subheadings (MeSH) and text words related to low grade glioma (LGG), molecular subtypes, and treatment, were utilized for the search strategy. Medline (PubMed interface, 2008 onwards), Embase (Ovid interface, 2008 onwards), and Cochrane Central Register for Controlled Trials (Wiley interface, current issue), were all searched. 1 January 2008 was selected as the start date for the search, based the first paper subcategorizing gliomas on the IDH molecular marker [11]. No other electronic database searches were conducted. In relevant literature, references were manually searched for additional trials. Search Strategy No limits, other than dates were utilized in the database search limitations. An electronic search examined Embase (January 1, 2008 to December 11, 2018), MEDLINE (January 1, 2008 to December 11, 2018), and Cochrane Central Register of Controlled Trails (CENTRAL) (January 1, 2008 to December 11, 2018); the Appendix provides the search protocols, including keywords. Specific search strategies were developed under guidance of Queen Square Institute of Neurology (IoN) library and statistical services staff with expertise in systemic review searches. To assess the search sensitivity and quality, robust target references were utilized—all of which were identified [6, 12, 13, 14, 15].

5

2.1.2 Study Records Data Management Results of the literature search were imported to EndNote X9 (Clarivate Analytics, Philadelphia, Pennsylvania). Software utilization aimed to reduce data entry errors and reduce bias, such as by deduplicating references. All reports on studies were reviewed, in order to assess for inconsistencies amongst reports (e.g. design description, outcome presentation, total patients analyzed). Selection Process The authors screened all titles and abstracts on the basis of the inclusion criteria. Subsequently, literature meeting inclusion criteria (including uncertain results) had the full-text reviewed. For results that met inclusion criteria, the literature was included in the systematic review. Data Items In accordance with recommendations from the Cochrane Handbook for Systematic Reviews of Interventions (chapter 7), the following data was collected into a Microsoft Excel spreadsheet: author, publication year, journal citation; setting; inclusion and exclusion criteria; study design; study population; tumor details at diagnosis (tumor size, location, and histology); risk of bias (including assessment of bias); length of follow-up; outcomes (OS, PFS) [16]. Data Synthesis Data were placed into tables allowing for relative comparison of OS and PFS stratified on tumor type and treatment. A quantile-quantile plot was produced for the PFS and OS data, which indicated both datasets to be non-normally distributed. Due to non-normal distribution when the data was pooled (cases with n=1 were excluded), the summary measures included the 25percentile, median, 75-percentile, and 95-percentile confidence interval of the median. Meanwhile, a nonparametric Kruskal-Wallis test was performed to determine if the survival outcomes stratified by genotype were significantly different; subsequently a pairwise analysis using the independent Wilcoxon rank sum test was utilized [17, 18]. All statistical analysis was conducted with R Statistical Software (R Foundation for Statistical Computing, Vienna, Austria) [19].

3.1 Results The search of Medline, Embase, and Central yielded a total of 8311 abstracts (Figure 1). Four additional papers were included after searching through systematic reviews identified in our search. After removing duplicates, 7542 abstracts were then screened by reading the title and full abstracts. From these, 7475 were excluded for not meeting the inclusion criteria of the study, and 67 were flagged for further review in the full-text assessment phase. Of the 67, 47 articles were removed for not providing raw data in the form of day/month/year for PFS or OS

6 (many abstracts met inclusion criteria, however provided data in the form of hazards ratios, pvalues, or Kaplan-Meier graphs without the ability to extract raw PFS or OS), or being systematic reviews—19 were included for quantitative synthesis in the form of Table 1 and Table 2. Of note, the systematic reviews that were excluded, did provide four additional articles that were also read in full, however none of these provided adequate data to be included in the quantitative synthesis. 3.1.1 Progression Free Survival Data From the 19 studies, 11 provided PFS data (Table 1). Six studies were retrospective and five were prospective (with one randomized); all studies looked at patients 18 years and older, except the Morshed et al. paper examined patients over 60 years old [6, 14, 20, 21, 22, 23, 24, 25, 26, 27, 28]. Baumert et al., in a randomized open label phase 3 intergroup study, examined conformal radiotherapy versus dose-dense temozolomide, in high risk tumors (Table 1) [6]. The Houillier et. al. retrospective study examined temozolomide administered daily for 5-days at 200mg/m , repeated every 28 days for at least 12 cycles (or up to 30 cycles) [22]. OD-II, A-mt II, and A-wt II had the following PFS, respectively: 37.9, 32.9, and 18.7 months [22]. On a dose-dense temozolomide regimen, 1 week on/1 week off, for a median of 11 cycles (range, 2-18 cycles), the prospective single arm phase II study by Pellerino et. al., found a PFS of 46 months for OD-II [26]. For temozolomide daily for 5-days at 200mg/day, repeated every 28 days up to 12 cycles, the prospective trial by Wahl et. al., found OD-II, A-mt II, and A-wt II to have the following PFS, respectively: 58.8, 43.2, and 7.2 months [14]. Finally, in a prospective phase II open label study, examining low-dose temozolomide 50mg/mq/day 1 week on/1 week off until progression (or for a maximum of 24 months), for OD-II and A-wt II, had a PFS of 35 and 6 months, respectively [27]. The retrospective study by Franceschi et al., examined the impact of postsurgical therapy on PFS for OD-II [21]. Regardless of treatment, this cohort had a PFS of 59.6 months, but with treatment PFS increased to 79.5 months [21]. When treatment was stratified, to follow-up only, CT alone, RT alone, and combined RT with CT, PFS incrementally increased: 46.3 months, 50.8 months, 103.6 months, and 120.2 months, respectively [21]. Rather than looking at OD-II, Minichillo et al. examined A-mt II, and how postsurgical treatment influences PFS; however, the data for postsurgical treatment with CT was not provided [23]. Without postsurgical treatment PFS was 44.3 months, but with any treatment was 64.8 months [23]. When treatment was stratified to follow-up only and RT, PFS was found to be 35.7 months and 60.0 months respectively [23]. Opoku-Darko et al. also provided data, but in the form of eight individual cases (Table 1) [25]. One additional paper also examined postsurgical treatment, comparing RT alone to RT with CT. High risk OD-II with RT and CT had a 162 months PFS, but with RT alone had a 91.2 months PFS [28].

7 When examining the impact of surgery across molecular subtypes, one retrospective study demonstrated OD-II, A-mt II, and A-wt II had a PFS of 109.2, 38.4, and 21 months, respectively [20]. When examining surgical resection outcomes stratified to age, for those older than 60 years, the following was found: OD-II, A-mt II, and A-wt II had a respective PFS of 37.3, 55.9, and 16.2 months [24]. After pooling all the data together on the basis of genotype a Kruskal-Wallis test found all three tumor types had significantly different median PFS (p = 0.01488), while a Wilcoxon ranked sum test found a pairwise comparison between all PFS of genotypes to also be significantly different (Table 2). The 25th percentile, median (50th percentile), 75th percentile, and 95% confidence interval of the median for PFS of each genotype was found, respectively—A-wt II: 6.90, 17.45, 19.57, and 3.00-23.69 months; A-mt II: 37.20, 43.20, 55.63, and 35.7-60.00 months; OD-II: 47.42, 59.20, 88.28, and 46.3-91.2 months (Table 2). A graphical representation stratified by treatment and genotype is displayed on Figure 2. 3.1.2 Overall Survival Data From the 19 included studies, 13 provided OS data (Table 2). Four studies were prospective and nine were retrospective [12, 13, 14, 15, 20, 24, 26, 28, 29, 30, 31, 32, 33]. Most of these studies examined surgical outcome per genotype, however several did look at other treatment modalities. Gao et al. noted for high risk A-wt II, postoperative conformal RT versus postoperative dose-dense oral TMZ had the respective median OS of 55.4 and 36 months [29]. On the other hand, A-wt II patients receiving radiotherapy alone or radiotherapy with chemotherapy (PCV), had an OS of 61.2 months [12]. Youland et al. on the other hand, for high risk OD-II, found for post-operative patients, RT with CT compared to RT alone, produced an OS of 212.4 versus 235.4 months, respectively [28]. Pellerino et al., for OD-II treated with dose-dense temozolomide 1 week on/1 week off for a median of 11 cycles (range 2-18 cycles), found an OS of 76 months [26]. Lastly, Wahl et al. examined patients with gross residual disease after surgical resection who received postoperative monthly temozolomide daily for 5-days at 200mg/day repeated every 28 days (up to 12 cycles), looking at OD-II, A-mt II, and A-wt II, the respective OS were 116.4, 134.4, and 21.6 months [14]. All other studies looked at outcomes per extent of resection. For instance, Jakola et al., noted for A-mt II, OS for patients with watchful waiting was 67.2 months, while 122.4 months for early resection; for A-wt II, OS was 16.8 months for watchful waiting and 63.6 months for early resection; no data was provided for oligodendrogliomas, as the data had yet to fully mature [13]. Likewise, when Wijnenga et al. examined surgical outcomes, the data had not yet matured for the OD-II category, but for A-mt II and A-wt II, those who underwent surgery had an OS of 122.4 and 25.2 months respectively [15]. However, Etxaniz et al. did present data for OD-II undergoing surgery. When comparing OD-II, A-mt II, and A-wt II, median OS was found to be 138, 115.2, and 22.8 months, respectively [20]. Looking at similar variables, Franceschi et al. noted for OD-II, A-mt II, and A-wt II undergoing surgery, OS was 216.0, 164.0, and 32.2 months, respectively [31]. Examining similar treatments,

8 another retrospective study found for OD-II without MGMT methylation, A-mt II with MGMT methylation, A-mt II without MGMT methylation, and A-wt II had the following respective OS: 189.4, 202.7, 109.1, 87.9 months [30]. Again, only one paper examined patients over the age of 60 years. Morshed et al. found for ODII, A-mt II, and A-wt II patients who underwent resection median OS was 67.3, 72.5, and 40.7 months, respectively [24]. After pooling all the data together on the basis of genotype a Kruskal-Wallis test found all three tumor types had significantly different median OS (p = 0.0001675), while a Wilcoxon ranked sum test found a pairwise comparison between OS for A-wt II versus A-mt II to be significantly different (Table 2). The 25th percentile, median (50th percentile), 75th percentile, and 95% confidence interval of the median for OS of each genotype was found, respectively—A-wt II: 22.8, 32.2, 40.7, and 21.6-61.2 months; A-mt II: 69.85, 115.2, 128.4, and 55.4-164.0; OD-II: 106.3, 163.7, 213.3, and 67.3-235.4 months (Table 2). A graphical representation stratified by treatment and genotype is displayed on Figure 3.

4.1 Discussion 4.1.1 Overall Survival—General Considerations Several key points can be extracted from this study. First, A-wt II tumors were found to have a significantly shorter OS compared to A-mt II (Table 2), confirming prior studies [20, 23, 34, 35]. Secondly, for OS, our data demonstrated no statistically significant difference between OD-II with A-mt II or A-wt II tumors (Table 2). Regarding PFS, all three tumor genotypes did have a statistically significant difference between them (Table 2). Finally, the PFS between A-mt II and OD-II was statistically significant (Table 2), with OD-II having a longer median PFS than A-mt II—such is potentially due to OD-II having earlier treatment with chemotherapy and radiotherapy than A-mt II tumors, and A-mt II tumors more likely to receive post-operative watch and scan treatment approach [36, 37]

4.2.1 PFS - WHO, grade II astrocytoma, Wild Type Several trends are identified when examining the raw data for PFS (Figure 2). For A-wt II patients, treatment with post-operative dose dense temozolomide (75mg/ daily for 21 days, repeated every 28 days for 12 cycles maximum) yielded the longest PFS, at 23.69 months [6]. The second longest PFS was with post-operative conformal RT, at 19.09 months [6]. Yet, there was no statistically significant treatment-dependent difference between the dose dense temozolomide and the conformal RT. Thus, to lengthen PFS, either dose dense temozolomide or conformal RT appear the best options. The dosages of temozolomide—dose dense regimen, standard schedule (200mg/ for 5 days, repeated every 28 days for 30 cycles maximum), and a low dose regimen (50mg/mq/day, 1

9 week on/1 week off)—were also compared. The dose dense regimen had the longest PFS at 23.69 months, followed by the standard schedule at 7.2 and 18.7 months, and finally the low dose regimen at 3 months [6, 14, 22, 25]. With the toxicity profile of dose dense and standard schedule relatively comparable, dose dense regimen appears more efficacious, yet due to the small number of studies, no safe conclusions can be made [38].

4.2.2 OS - WHO, grade II astrocytoma, Wild Type However, for prolonging OS, a different treatment regimen was more beneficial—early resection with radiotherapy [12, 13]. When comparing watchful waiting to early resection, OS was 16.8 and 63.6 months, respectively [13]. Now when examining post-operative treatments, a radiotherapy-based regimen (RT alone or with PCV) resulted in the longest OS, at 61.2 months (post-operative temozolomide yielded in 21.6 months) [12, 14]. Therefore, regarding OS, a post-operative radiotherapy-based treatment regimen after early resection seems most beneficial for A-wt II tumors (Figure 3).

4.3.1 PFS - WHO, grade II astrocytoma, Mutant Amongst A-mt II patients, radiotherapy had a better PFS than treatment with temozolomide or post-surgical follow-up only (Figure 2). Two studies provided PFS for treatment with radiotherapy: 60 and 55.36 months [6, 23]. The three temozolomide data points did not demonstrate any observable trends with regards to dosage, rather the PFS were: 32.9, 36.1, and 43.2 months [6, 14, 22]. Post-surgical follow-up only had a value of 35.7 months, falling within the extremes of the temozolomide PFS values [23]. With the data currently available, to maximize PFS, radiotherapy appears to yield the best results for A-mt II. Overall, further studies with large sample sizes and directly comparing various treatment modalities are needed.

4.3.2 OS - WHO, grade II astrocytoma, Mutant The dataset for A-mt II was composed of eleven values, with none of the studies providing conclusive insight into treatment (Figure 3). For instance, one study indicated for high risk A-wt, RT (55.4 months) had a significantly longer OS than dose-dense TMZ (36 months) [29]. Meanwhile, comparing across two different studies, treatment with temozolomide (134.4 months) provided a longer OS than watchful waiting (67.2 months) [13, 14]. Thus, further studies examining the nuances of the treatment modalities are needed for A-mt II prior to making conclusions.

4.4.1 PFS - WHO, grade II oligodendroglioma For OD-II, combination RT with CT produces the longest PFS, followed by RT alone, and finally temozolomide alone (Figure 2). A combination of radiotherapy and chemotherapy was found to have PFS at 120.2 and 162 months. Meanwhile, radiotherapy had the second longest PFS at 61.63, 91.2, and 103.6 months [6, 21, 28]. Finally, studies of temozolomide treatment alone produced the shortest PFS values: 35, 37.9, 46, 55.03, and 58.8 months [6, 14, 21, 22, 26, 27].

10 Therefore, on the basis of this data, for those with OD-II a combination of RT and CT should be considered, yet more robust comparative prospective clinical trials are needed.

4.4.2 OS - WHO, grade II oligodendroglioma Amongst OD-II tumors, OS values stratified by treatment were inconclusive (Figure 3). One study examined radiotherapy, providing the longest OS value of all studies at 235.4 months [28]. Following the single RT value of 235.4, the second longest OS was with a combination of RT and CT (212.4 months) [28]. For temozolomide, OS values were 76 and 116.4 months, which was conclusively distinguishable from radiotherapy (235.4 months) [14, 26, 28]. Overall, RT with CT or RT alone maximizing OS for OD-II; however, better data will be needed before any conclusions can be drawn.

4.5.1 Stratification by Age When stratifying based on age, those older than 60 years had different trends. For the general population without age-stratification, several of the datasets highlighted how OD-II have better survival outcomes than A-mt II, and A-mt II have better outcomes than A-wt II. Etxaniz et al. and Opoku-Darko et al. demonstrated this trend with surgery, Houillier et al. with postoperative initial TMZ treatment, and Wahl et al. with postoperative monthly TMZ cycles [14, 20, 22, 25]. However, these trends change when age is stratified. The data from Morshed et al. studied patients older than 60 years and found A-mt II tumors responded better to therapy than OD-II, and OD-II better than A-wt II [24]. Hence, future study designs should stratify results based on age—medicine well recognizes geriatric patients as having many physiologic differences compared to younger adults, such as liver metabolism amongst others [39].

4.6.1 Personalized treatments Future optimal management strategies are likely to involve personalized paradigms. Even within molecular subtypes, each patient will likely have numerous confounding variables that influence outcomes. When looking at the case series of eight patients in the Opoku-Darko et al. study (Table 1), some of the earlier observed trends do not conform to this case series [25]. For instance, the OD-II tumor with the greatest EOR had a PFS less than the patient with biopsy, while one A-mt II tumor with 100% EOR had a PFS less than a biopsy patient, and another A-mt II tumor with chemoradiotherapy had a worse PFS outcome than a patient who did not receive RT or CT [25]. Therefore, the likely end result for LGG management will possibly be a complex algorithm that takes into consideration an abundant array of individualized patient variables, including possibly more genetic markers.

4.7.1 Study Limitations There are a number of limitations to this systematic review and the included studies. For example, several datasets implied rather than providing explicit definitions for PFS and OS. Moreover, others did not detail the CT or RT regimen details, or simply combined CT and RT

11 together. Likewise, several of the methods sections minimally described the inclusion and exclusion criteria. Regarding statistics, certain datasets did not provide confidence intervals. In addition, most studies included were retrospective, while some studies essentially compared patients who were suitable surgical candidates versus unsuitable (as surgical randomization is untenable due to lack of clinical equipoise). Another limitation is the inconsistent follow-up times amongst the studies, ranging from 20 to 216 months (18 years). Lastly, with this systematic review, most of the included studies involved small samples sizes, therefore limiting the conclusions. However, by presenting survival outcome data, this systematic review also highlights the problems with the ability to draw conclusions from current studies, thus by addressing these problems future clinical trial design may improve. However, by extension no treatment recommendations can be made from this review.

5.1 Conclusion Overall, after stratifying WHO grade II gliomas based on molecular subtype and treatment modality, several findings were made regarding OS and PFS. Median OS for A-wt II (32.2 months), A-mt II (115.2 months), and OD-II (163.7 months) were found, as were median PFS for A-wt II (17.45 months), A-mt (43.20 months), and OD-II (59.20 months). Overall, A-wt II was confirmed to have a significantly shorter OS than A-mt II; meanwhile, there was no significant difference found between OS of OD II with A-mt II and A-wt II. Additionally, all three molecular subtypes were found to have statistically significant differences between PFS, with OD II having a statistically better PFS than A-mt II. Future studies should also attempt to examine quality of life variables, accounting for treatment toxicity or post-operative neurologic function loss. However, knowing the raw PFS or OS data helps design better future clinical trials and provides patients with more tangible information when making critical and complex treatment decisions. Additionally, despite the limitations, this study makes sense of the best available data, helping better elucidate the progression and outcomes of WHO grade II gliomas.

12

Compliance with Ethical Standards Declarations of Interest: None Author Arash Ghaffari-Rafi declares he has no conflict of interest. Author George Samandouras declares he has no conflict of interest Ethical Approval This article does not contain any studies with human participants or animals performed by any of the authors.

Acknowledgements We would like to thank Dr. Saiful Islam from the UCL Queens Square Institute of Neurology Office of Statistics for his time in providing valuable statistical consultation and Dr. Matthew Kirkman for stylistically reviewing the manuscript.

13

References [1] Pouratian N, Schiff D. Management of Low-Grade Glioma. Curr Neurol Neurosci Rep. 2010;10(3):224-231 [2] Jiang B, Chaichana K, Veeravagu A, Chang SD, Black KL, Patil CG. Biopsy versus resection for the management of low-grade gliomas. Cochrane Database Syst Rev. 2017;2017(4) [3] Grier J. Low-Grade Gliomas in Adults. The Oncologist. 2006;11(6):681-693. [4] Whittle IR The dilemma of low grade glioma Journal of Neurology, Neurosurgery & Psychiatry 2004;75:ii31-ii36. [5] Louis, D., Ohgaki, H., Wiestler, O. and Cavenee, W. 2016. WHO classification of tumours of the central nervous system. 4th ed. WHO/IARC. [6] Baumert BG, Hegi ME, van den Bent MJ, von Deimling A, Gorlia T, Hoang-Xuan K, et al. Temozolomide chemotherapy versus radiotherapy in high-risk low-grade glioma (EORTC 2203326033): a randomised, open-label, phase 3 intergroup study. The Lancet Oncology. 2016;17(11):1521-32. [7] Higgins, J., Altman, D., Gotzsche, P., Juni, P., Moher, D., Oxman, A., Savovic, J., Schulz, K., Weeks, L. and Sterne, J. (2011). The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ, 343(oct18 2), pp.d5928-d5928. [8] Moher, D., Shamseer, L., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., Shekelle, P. and Stewart, L. 2015. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Systematic Reviews, 4(1). [9] Shamseer, L., Moher, D., Clarke, M., Ghersi, D., Liberati, A., Petticrew, M., Shekelle, P. and Stewart, L. 2015. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ, 349(Jan), pp.g7647-g7647. [10] Reuss DE, Sahm F, Schrimpf D, et al. ATRX and IDH1-R132H immunohistochemistry with subsequent copy number analysis and IDH sequencing as a basis for an “integrated” diagnostic approach for adult astrocytoma, oligodendroglioma and glioblastoma. Acta Neuropathol. 2015;129(1):133-146. [11] Parsons, D. W., Jones, S., Zhang, X., Lin, J. C., Leary, R. J., Angenendt, P., Mankoo, P., Carter, H., Siu, I. M., Gallia, G. L., Olivi, A., McLendon, R., Rasheed, B. A., Keir, S., Nikolskaya, T., Nikolsky, Y., Busam, D. A., Tekleab, H., Diaz, L. A., Hartigan, J., Smith, D. R., Strausberg, R. L., Marie, S. K., Shinjo, S. M., Yan, H., Riggins, G. J., Bigner, D. D., Karchin, R., Papadopoulos, N.,

14 Parmigiani, G., Vogelstein, B., Velculescu, V. E., Kinzler, K. W. 2008. An integrated genomic analysis of human glioblastoma multiforme. Science, 321(5897), 1807-12. [12] Buckner J, Giannini C, Eckel-Passow J, Lachance D, Parney I, Laack N, et al. Management of diffuse low-grade gliomas in adults - use of molecular diagnostics. Nature reviews Neurology. 2017;13(6):340-51. [13] Jakola, A., Skjulsvik, A., Myrmel, K., Sjåvik, K., Unsgård, G., Torp, S., Aaberg, K., Berg, T., Dai, H., Johnsen, K., Kloster, R. and Solheim, O. 2017. Surgical resection versus watchful waiting in low-grade gliomas. Annals of Oncology, 28(8), pp.1942-1948. [14] Wahl M, Phillips JJ, Molinaro AM, Lin Y, Perry A, Haas-Kogan DA, et al. Chemotherapy for adult low-grade gliomas: clinical outcomes by molecular subtype in a phase II study of adjuvant temozolomide. Neuro-oncology. 2017;19(2):242-51. [15] Wijnenga MMJ, French PJ, Dubbink HJ, W.N.M DI, Atmodimedjo PN, Kros JM, et al. The impact of surgery in molecularly defined low-grade glioma: An integrated clinical, radiological, and molecular analysis. Neuro-oncology. 2018;20(1):103-12. [16] Higgins, J., Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration 2011. Available from www.cochranehandbook.org. [17] William H. Kruskal & W. Allen Wallis (1952) Use of Ranks in One-Criterion Variance Analysis, Journal of the American Statistical Association, 47:260, 583621, DOI: 10.1080/01621459.1952.10483441 [18] Wilcoxon, Frank. "Individual Comparisons by Ranking Methods." Biometrics Bulletin 1, no. 6 (1945): 80-83. doi:10.2307/3001968. [19] R Core Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org/. [20] Etxaniz O, Carrato C, de Aguirre I, Queralt C, Munoz A, Ramirez JL, et al. IDH mutation status trumps the Pignatti risk score as a prognostic marker in low-grade gliomas. Journal of neuro-oncology. 2017;135(2):273-84. [21] Franceschi E, De Biase D, Paccapelo A, Mura A, Tosoni A, Bartolini S, et al. Low grade glioma patients with IDH mutation and 1p19q codeletion: To treat or not to treat? Journal of Clinical Oncology Conference. 2017;35(15 Supplement 1). [22] Houillier C, Wang X, Kaloshi G, Mokhtari K, Guillevin R, Laffaire J, et al. IDH1 or IDH2 mutations predict longer survival and response to temozolomide in low-grade gliomas. Neurology. 2010;75(17):1560-6.

15

[23] Minichillo S, Franceschi E, Mura A, Tosoni A, Tallini G, Pession A, et al. The role of treatments in IDH mutant molecular astrocytomas. Annals of Oncology. 2017;28 (Supplement 6):vi76. [24] Morshed RA, Han SJ, Hervey-Jumper SL, Pekmezci M, Troncon I, Chang SM, et al. Molecular features and clinical outcomes in surgically treated low-grade diffuse gliomas in patients over the age of 60. Journal of neuro-oncology. 2018. [25] Opoku-Darko M, Lang ST, Artindale J, Cairncross JG, Sevick RJ, Kelly JJP. Surgical management of incidentally discovered diffusely infiltrating low-grade glioma. Journal of neurosurgery. 2018;129(1):19-26. [26] Pellerino A, Franchino F, Pace A, Carapella C, Dealis C, Caroli M, et al. Temozolomide (TMZ) 1 week on/1 week off as initial treatment for high risk low grade oligodendroglial tumors: A phase II aino (Italian association for neuro-oncology) study. Neuro-oncology. 2017;19 (Supplement 3):iii21. [27] Villani V, Merola R, Vidiri A, Fabi A, Carosi M, Giannarelli D, et al. Temozolomide low-dose chemotherapy in newly diagnosed low-grade gliomas: activity, safety, and long-term follow-up. Tumori. 2017;103(3):255-60. [28] Youland RS, Kreofsky CR, Schomas DA, Brown PD, Buckner JC, Laack NN. The impact of adjuvant therapy for patients with high-risk diffuse WHO grade II glioma. Journal of neurooncology. 2017;135(3):535-43. [29] Gao Y, Weenink B, van den Bent MJ, Erdem-Eraslan L, Kros JM, Sillevis Smitt PAE, et al. Expression-based intrinsic glioma subtypes are prognostic in low-grade gliomas of the EORTC22033-26033 clinical trial. European Journal of Cancer. 2018;94:168-78. [30] Brandes AA, Paccapelo A, De Blase D, Reni M, Mura A, Bartolini S, et al. The role of clinical characteristics and molecular biomarkers in low grade gliomas (LGG): A GICNO study. Journal of Clinical Oncology Conference. 2016;34(Supplement 15). [31] Franceschi E, Mura A, De Biase D, Tallini G, Pession A, Foschini MP, et al. The role of clinical and molecular factors in low-grade gliomas: what is their impact on survival? Future oncology (London, England). 2018;14(16):1559-67. [32] Poulen G, Goze C, Rigau V, Duffau H. Huge heterogeneity in survival in a subset of adult patients with resected, wild-type isocitrate dehydrogenase status, WHO grade II astrocytomas. Journal of neurosurgery. 2018:1-10. [33] Yeboa D, Yu J, Huse J, Penas-Prado M, Sulman E, Contessa J. Patterns of treatment and outcomes of 1P19Q co-deleted gliomas. Neuro-oncology. 2017;19 (Supplement 6):vi222.

16

[34] Li M-Y, Wang Y-Y, Cai J-Q, Zhang C-B, Wang K-Y, Cheng W, Liu Y-W, Zhang W, Jiang T (2015) Isocitrate dehydrogenase 1 gene mutation is associated with prognosis in clinical low-grade gliomas. PLoS One 10:e0130872. https://doi.org/10.1371/journal.pone.0130872 [35] Metellus P, Coulibaly B, Colin C, De Paula AM, Vasiljevic A, Taieb D, Barlier A, Boisselier B, Mokhtari K, Wang XW, Loundou A, Chapon F, Pineau S, Ouafik L, Chinot O, FigarellaBranger D (2010) Absence of IDH mutation identifies a novel radiologic and molecular subtype of WHO grade II gliomas with dismal prognosis. Acta Neuropathol 120:719– 729. https://doi.org/10.1007/s00401-010-0777-8 [36] Straube C, Kessel KA, Schmidt-Graf F, et al. A trend towards a more intense adjuvant treatment of low-grade-gliomas in tertiary centers in Germany after RTOG 9802 - results from a multi-center survey. BMC Cancer. 2018;18(1):907. Published 2018 Sep 21. doi:10.1186/s12885-018-4825-4 [37] van den Bent MJ. Oligodendrogliomas: a short history of clinical developments. CNS Oncol. 2015;4(5):281–285. doi:10.2217/cns.15.35 [38] Pouratian, N., Gasco, J., Sherman, J. H., Shaffrey, M. E., Schiff, D. 2007. Toxicity and efficacy of protracted low dose temozolomide for the treatment of low grade gliomas. J Neurooncol, 82, 281-8. [39] Mangoni, A. A., & Jackson, S. H. 2004. Age-related changes in pharmacokinetics and pharmacodynamics: basic principles and practical applications. British journal of clinical pharmacology, 57(1), 6-14.

17

Appendix Pubmed (MEDLINE) Search Strategy (A) (1) (2) (3) (4)

molecula* genetic* or genetics or genetic mutation* or mutation molecular genetic* or molecular genetic or molecular genetics

Search (((molecula*) OR (((genetic*) OR genetics) OR genetic)) OR ((mutation*) OR mutation)) OR (((molecular genetics) OR molecular genetic) OR molecular genetic*) (B) (1) overall survival* or overall survival or overall survivals (2) survival* or survival or survivals (3) “os” Search (((((((overall survival*) OR overall survival) OR overall survivals) OR survival*) OR survival) OR survivals) OR "os")

(C) (1) progression free survival* or progression free survival or progression free survivals (2) progression* or progression or progressions or PFS or PFSs Search ((((((((progression free survival*) OR progression free survival) OR progression free survivals) OR progression*) OR progression) OR progressions) OR PFS) OR PFSs) (1) (2) (3) (4)

low grade glioma or LGG or LGGs grade 2 gliomas or grade ii gliomas astrocytoma* or astrocytomas oligodendroglioma* or oligodendrogliomas

Search ((((((low grade glioma) OR LGG) OR LGGs)) OR ((grade 2 gliomas) OR grade ii gliomas)) OR ((astrocytoma*) OR astrocytomas)) OR ((oligodendroglioma*) OR oligodendrogliomas)

(D) (1) treatment* or treatments or treatment (2) treat* or treat or treats Search (((((treatment*) OR treatments) OR treatment) OR treat*) OR treat) OR treats

18

Sort by: Best Match Filters: Publication date from 2008/01/01 to 2018/12/31

Search (((((((((((((overall survival*) OR overall survival) OR overall survivals) OR survival*) OR survival) OR survivals) OR "os"))) OR (((((((((progression free survival*) OR progression free survival) OR progression free survivals) OR progression*) OR progression) OR progressions) OR PFS) OR PFSs)))) AND ((((molecula*) OR (((genetic*) OR genetics) OR genetic)) OR ((mutation*) OR mutation)) OR (((molecular genetics) OR molecular genetic) OR molecular genetic*))) AND (((((((low grade glioma) OR LGG) OR LGGs)) OR ((grade 2 gliomas) OR grade ii gliomas)) OR ((astrocytoma*) OR astrocytomas)) OR ((oligodendroglioma*) OR oligodendrogliomas))) AND ((((((treatment*) OR treatments) OR treatment) OR treat*) OR treat) OR treats)

19

Embase Ovid Search Strategy (1) (2) (3) (4) (5) (6) (7)

exp glioma/ glioma*.mp. LGG*.mp. astrocytoma*.mp. oligodendroglioma*.mp. (grade adj ii).mp. 1 or 2 or 3 or 4 or 5 or 6

(8) exp progression free survival/ (9) (progression adj free).mp. (10) progression*.mp. (11) PFS*.mp. (12) 8 or 9 or 10 or 11 (13) (14) (15) (16)

exp overall survival (overall adj survival*).mp. OS*.mp. 13 or 14 or 15

(17) (18) (19) (20) (21)

exp molecular genetics (molecular adj genetic*).mp. molecul*.mp. genetic*.mp. 17 or 18 or 19 or 20

(22)

treatment*.mp.

(23) (24) (25)

12 or 16 7 and 21 and 22 and 23 24 and 2008:2018.(sa_year).

Key: mp = title, abstract, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword, floating subheading word, candidate term wor

20

CENTRAL Search Strategy (1) (2) (3) (4) (5) (6) (7)

MeSH descriptor: [Glioma] explode all trees glioma* astrocytoma* oligodendroglioma* LGG* #2 or #3 or #4 or #5 # 1 or #6

(8) MeSH descriptor: [Disease-Free Survival] explode all trees (9) progression* (10) survival* (11) PFS* (12) #9 or #10 or #11 (13) #8 or #12 (14) (15) (16)

OS* overall* #14 or #15

(17)

#13 or #16

(18) (19) (20) (21)

MeSH descriptor: [Molecular Biology] explode all trees genetic* molecul* #18 or #19 or #20

(22) (23) (24)

MeSH descriptor: [Therapeutics] explode all trees treatment* #22 or #23

(25)

#7 and #17 and #21 and #24

Table 1: PFS and OS Stratified by Tumor Molecular Subtype and Treatment Study

Baumert et al., 2016 [6]

Etxaniz et al., 2017 [20]

Median Follow-Up (months)

n

Molecular Subtype

Treatment

Median PFS (months)

89

Astrocytoma-mutant II, High Risk

Radiotherapy

55.36 (95% CI: 47.87-65.87)

76

Astrocytoma-mutant II, High Risk

Temozolomide (dose dense,75mg/m2 daily, for 21 days, repeated every 28 days, for 12 cycles maximum)

36.01 (95% CI: 28.42-46.95)

45

Oligodendroglioma II, High Risk

Radiotherapy

61.63 (95% CI: 42.32-NR)

59

Oligodendroglioma II, High Risk

Temozolomide (dose dense,75mg/m2 daily, for 21 days, repeated every 28 days, for 12 cycles maximum)

55.03 (95% CI: 37.95-NR)

29

Astrocytoma-wild type II, High Risk

Radiotherapy

19.09 (95% CI: 11.27-25.69)

20

Astrocytoma-wild type II, High Risk

Temozolomide (dose dense,75mg/m2 daily, for 21 days, repeated every 28 days, for 12 cycles maximum)

23.69 (95% CI: 5.55-42.25)

14

Oligodendroglioma II

Surgery or Biopsy +/- Postsurgical Treatment

109.2 (95% CI: 78-138)

11

Astrocytoma-mutant II

Surgery or Biopsy +/- Postsurgical Treatment

38.4 (95% CI: 0-99.6)

12

Astrocytoma-wild type II

Surgery or Biopsy +/- Postsurgical Treatment

21 (95% CI: 0-9.2)

93

Oligodendroglioma II

Surgery or Biopsy +/- Postsurgical Treatment

59.6 (95% CI: 41.8-77.4)

43

Oligodendroglioma II

Surgery or Biopsy + Postsurgical Treatment (CT/RT/both)

79.5 (95% CI: 66.4-92.7)

50

Oligodendroglioma II

Surgery or Biopsy + Follow-Up Only

46.3 (95% CI: 36.0-56.5)

17

Oligodendroglioma II

Chemotherapy

50.8 (95% CI: 17.4-84.3)

18

Oligodendroglioma II

Radiotherapy

103.6 (95% CI: 11.7-195.6)

8

Oligodendroglioma II

Radiotherapy and Chemotherapy

120.2 (95% CI: 40.5-199.8)

Oligodendroglioma II

Temozolomide (daily for 5 days, at 200mg/m2, repeated every 28 days, at least 12 cycles or up to 30)

37.9

Astrocytoma-mutant II

Temozolomide (daily for 5 days, at 200mg/m2, repeated every 28 days, at least 12 cycles or up to 30)

32.9

Astrocytoma-wild type II

Temozolomide (daily for 5 days, at 200mg/m2, repeated every 28 days, at least 12 cycles or up to 30)

18.7

48 months

91.2 months

Median PFS (3-Years %)

HR 1.86 (95% CI, 1.21-2.87), log-rank p = 0.0043

No Significant treatmentdependent differences

No Significant treatmentdependent differences

P = 0.003

p = 0.001 Franceschi et al., 2017 [21]

Houillier et al., 2010 [22]

96.1 months

63.4 months

74

These differences did not reach significance

Median PFS (5-Years %) 42.50% (95% CI: 27.3856.83) 19.43% (95% CI: 8.87-33.00)

58.49% (95% CI: 39.4373.41) 47.39% (95% CI: 30.7162.35)

0% (95% CI: 0.00-0.00) 17.78% (95% CI: 3.69-40.48)

Minichillo et at., 2017 [23]

Morshed et al., 2018 [24]

Opoku-Darko et al., 2018 [25]

Pellerino et al., 2017 [26]

Wahl et al., 2017 [14]

98.3 months

62.4 months

50.4 months

90

Astrocytoma-mutant II

Surgery or Biopsy +/- Postsurgical Treatment

44.3

31

Astrocytoma-mutant II

Surgery or Biopsy + Postsurgical Treatment (CT/RT/both)

64.8

59

Astrocytoma-mutant II

Surgery or Biopsy + Follow-Up Only

35.7

20

Astrocytoma-mutant II

Radiotherapy

60

5

Astrocytoma-mutant II

Surgery (patients at least 60 years old) +/- Postsurgical Treatment

55.9 (95% CI: 32.7-79.1)

7

Astrocytoma-wild type II

Surgery (patients at least 60 years old) +/- Postsurgical Treatment

16.2 (95% CI: 0-35.8)

13

Oligodendroglioma II

Surgery (patients at least 60 years old) +/- Postsurgical Treatment

37.3 (95% CI: 22.9-51.7)

3

Oligodendroglioma II

Surgery or Biopsy + Postsurgical Treatment (CT/RT/both)

67 (for 105, 67, and 40)

1

Oligodendroglioma II

Biopsy + CT and RT

105

1

Oligodendroglioma II

86.42% EOR; CT + RT

67

1

Oligodendroglioma II

68.98% EOR; RT

40

4

Astrocytoma-mutant II

Surgery or Biopsy +/- Postsurgical Treatment

38.5 (for 16, 32, 45, and 48)

1

Astrocytoma-mutant II

100% EOR; no repeat resection;

32

1

Astrocytoma-mutant II

51.82% EOR; no repeat resection; RT + CT

16

1

Astrocytoma-mutant II

76.28% EOR; repeat resection;

45

1

Astrocytoma-mutant II

biopsy; repeat resection

48

1

Astrocytoma-wild type II

Surgery (87.77% EOR; no repeat resection)

3

21

Oligodendroglioma II

Temozolomide (dose dense, 1 week on/1 week off, median 11 cycles, range 2-18 cycles)

46

71.40%

28.60%

24

IDH-wild type

Temozolomide (dose dense, 1 week on/1 week off, median 11 cycles, range 2-18 cycles)

34

45.80%

25.00%

44

Oligodendroglioma II (imputed)

Temozolomide (daily for 5 days, at 200mg/day, repeated every 28 days, up to 12 cycles)

58.8 (95% CI: 45.6-NA)

31

Astrocytoma-mutant II

Temozolomide (daily for 5 days, at 200mg/day, repeated every 28 days, up to 12 cycles)

43.2 (95% CI: 25-64.8)

13

Astrocytoma-wild type II

Temozolomide (daily for 5 days, at 200mg/day, repeated every 28 days, up to 12 cycles)

7.2 (95% CI: 4.8-NA)

p = 0.004

p = 0.004, for treated with RT versus Follow-Up

p = 0.04

No p-values provided

64 months

90 months

p = 0.01

Oligodendroglioma II (imputed)

Temozolomide (low dose, 50mg/mq/day, 1 week on/1 week off, until progression or for maximum of 24 months)

35

2

No 1p19q Codeletion

Temozolomide (low dose, 50mg/mq/day, 1 week on/1 week off, until progression or for maximum of 24 months)

5

7

IDH-mutant

Temozolomide (low dose, 50mg/mq/day, 1 week on/1 week off, until progression or for maximum of 24 months)

36

4

Astrocytoma-wild type II

Temozolomide (low dose, 50mg/mq/day, 1 week on/1 week off, until progression or for maximum of 24 months)

6

15

Oligodendroglioma II (imputed), High Risk

Radiotherapy and Chemotherapy

162

29

Oligodendroglioma II (imputed), High Risk

Radiotherapy

91.2

10

p < 0.04 Villani et al., 2017 [27]

54 months

p < 0.009

Youland et al., 2017 [28]

Study

Median Follow-Up (months)

Gao et al., 2018 [29]

48 months

127.2 months

p = 0.046

n

Molecular Subtype

Treatment

Median OS (months)

41

Astrocytoma-mutant II, High Risk

Temozolomide (dose dense,75mg/m2 daily, for 21 days, repeated every 28 days, for 12 cycles maximum)

36

51

Astrocytoma-mutant II, High Risk

Radiotherapy

55.4

Oligodendroglioma II (imputed)

Surgery or Biopsy +/- Postsurgical Treatment

186.2 (95% CI: 128.2-244.2)

Median OS (2-Years %)

Median OS (3-Years %)

Median OS (5-Years %)

Median OS (10-Years %)

94.70%

73.80%

p < 0.01

p = 0.03 Non 1p19q Codeletion

Surgery or Biopsy +/- Postsurgical Treatment

138.5 (95% CI: 91.1-186.0)

84.90%

51.50%

Oligodendroglioma II + MGMT Methylation

Surgery or Biopsy +/- Postsurgical Treatment

NA

100%

95.70%

Oligodendroglioma II + No MGMT Methylation

Surgery or Biopsy +/- Postsurgical Treatment

189.4

92.90%

61.90%

Astrocytoma-mutant II + MGMT Methylation

Surgery or Biopsy +/- Postsurgical Treatment

202.7

95.80%

72.90%

Astrocytoma-mutant II + No MGMT Methylation

Surgery or Biopsy +/- Postsurgical Treatment

109.1 (95% CI: 105.3-112.8)

90.70%

44.10%

NA

Astrocytoma-wild type II

Surgery or Biopsy +/- Postsurgical Treatment

87.9 (95% CI: 61.1-114.6)

20

Astrocytoma-wild type II

Radiotherapy Alone or Radiotherapy with Chemotherapy (PCV)

61.2 (95% CI: 22.8-138)

36

IDH1-mutant

Radiotherapy Alone or Radiotherapy with Chemotherapy (PCV)

157.2

14

Oligodendroglioma II

Surgery or Biopsy +/- Postsurgical Treatment

138 (95% CI: 32.4-243.6)

11

Astrocytoma-mutant II

Surgery or Biopsy +/- Postsurgical Treatment

115.2 (95% CI: 108-122.4)

198 Brandes et al., 2016 [30]

Buckner et al., 2017 [12]

Etxaniz et al., 2017 [20]

74.0 months

142.8 months

p < 0.001 (for IDHmt vs IDHwt)

p = 0.02

91.2 months

p < 0.001

12

Astrocytoma-wild type II

Surgery or Biopsy +/- Postsurgical Treatment

22.8 (95% CI: 0-60)

198

IDH-mutant

Surgery or Biopsy +/- Postsurgical Treatment

187.2 (95% CI: 148.0-226.4)

15

Astrocytoma-wild type II

Surgery or Biopsy +/- Postsurgical Treatment

32.2 (95% CI: 0.0-70.357)

96

Oligodendroglioma II (imputed)

Surgery or Biopsy +/- Postsurgical Treatment

189.4 (95% CI: 154.5-224.346)

101

Non 1p19q Codeletion

Surgery or Biopsy +/- Postsurgical Treatment

164.0 (95% CI: 96.6-231.5)

108

Oligodendroglioma II

Surgery or Biopsy +/- Postsurgical Treatment

216.0 (95% CI: 177.6-254.3)

98.4 +/- 1.6% 94.4 +/- 2.7% 49.9 +/13.6%

p < 0.001

p = 0.015

Franceschi et al., 2018 [31]

98.3 months

90

Astrocytoma-mutant II

Surgery or Biopsy +/- Postsurgical Treatment

164.0 (95% CI: 100.3-227.7)

15

Astrocytoma-wild type II

Surgery or Biopsy +/- Postsurgical Treatment

32.2 (95% CI: 15.4-0.0)

163

IDH-mutant, High Risk

Surgery or Biopsy +/- Postsurgical Treatment

164

13

Astrocytoma-wild type II, High Risk

Surgery or Biopsy +/- Postsurgical Treatment

32.3

93

Oligodendroglioma II (imputed), High Risk

Surgery or Biopsy +/- Postsurgical Treatment

189.4

83

Non 1p19q Codeletion, High Risk

Surgery or Biopsy +/- Postsurgical Treatment

138.5

43

Oligodendroglioma II

Watchful Waiting vs. Early Resection

Not Reached

25

Astrocytoma-mutant II

Watchful Waiting

67.2 (95% CI: 42-91.2)

36

Astrocytoma-mutant II

Early Resection

122.4 (95% CI: 82.8-160.8)

16

Astrocytoma-wild type II

Watchful Waiting

16.8 (95% CI: 7.2-26.4)

25

Astrocytoma-wild type II

Early Resection

63.6 (95% CI: 0.0-240.0)

5

Astrocytoma-mutant II

Surgery (patients at least 60 years old) +/- Postsurgical Treatment

72.5 (95% CI: 40.9-104.1)

7

Astrocytoma-wild type II

Surgery (patients at least 60 years old) +/- Postsurgical Treatment

40.7 (95% CI: 14.1-67.4)

13

Oligodendroglioma II

Surgery (patients at least 60 years old) +/- Postsurgical Treatment

67.3 (95% CI: 47.7-86.9) 76

213

p < 0.001

p = 0.015

Jakola et al., 2017 [13]

Morshed et al., 2018 [24]

Range: 84216 months

62.4 months

Pellerino et al., 2017 [26]

64 months

15

Oligodendroglioma II

Temozolomide (dose dense, 1 week on/1 week off, median 11 cycles, range 2-18 cycles)

Poulen et al., 2018 [32]

74 months

31

Astrocytoma-mutant II

Surgery without Early Adjuvant

p = 0.21

90.50%

66.70%

77.27%

Wahl et al., 2017 [14]

Wijnenga et al., 2018 [15]

Youland et al., 2017 [28]

Yeboa et al., 2017 [33]

90 months

69.48 months

44

Oligodendroglioma II

Temozolomide (daily for 5 days, at 200mg/day, repeated every 28 days, up to 12 cycles)

116.4 (95% CI: 87.6-NA)

31

Astrocytoma-mutant II

Temozolomide (daily for 5 days, at 200mg/day, repeated every 28 days, up to 12 cycles)

134.4 (95% CI: 82.8-NA)

13

Astrocytoma-wild type II

Temozolomide (daily for 5 days, at 200mg/day, repeated every 28 days, up to 12 cycles)

21.6 (95% CI: 13.2-NA)

23

Astrocytoma-wild type II

Surgery or Biopsy +/- Postsurgical Treatment

25.2

112

Astrocytoma-mutant II

Surgery or Biopsy +/- Postsurgical Treatment

122.4

93

Oligodendroglioma II

Surgery or Biopsy +/- Postsurgical Treatment

Not Reached

15

Oligodendroglioma II (imputed), High Risk

Radiotherapy and Chemotherapy

212.4

29

Oligodendroglioma II (imputed), High Risk

Radiotherapy

235.4

472

Oligodendroglioma II (imputed)

Surgery

127.2 months

NA

p < 0.001

p = 0.0001

p = 0.046

99.30%

Table 2: Interquartile Range, Median, Confidence Interval for PFS and OS Stratified by Tumor Molecular Subtype th

th

OS 25 Quartile (months)

OS Median (months)

OS 75 Quartile (months)

OS 95% Confidence Interval of Median (months)

Astrocytoma-wild type II

22.8

32.2

40.7

(21.6, 61.2)

Astrocytoma-mutant II

69.85

115.2

128.4

(55.4, 164.0)

Oligodendroglioma II

106.3

163.7

213.3

(67.3, 235.4)

th

Astrocytoma-wild type II

Wilcoxon Ranked Sum Test (Pairwise Comparison) Kruskal-Wallis Test Astrocytoma-mutant II

Oligodendroglioma II

76.9 (95% CI: 34.9-102.2), p = 0.0008255

123.5 (95% CI: 52.8-183.8), p = 0.055433 51.3 (95% CI: -13.3-113.0), p = 0.09037

31.5 (95% CI: 19.5-41.11), p = 0.01783

45.1 (95% CI: 32.1-75.0), p = 0.0031728

17.389 (p-value = 0.0001675)

th

PFS 25 Quartile (months)

PFS Median (months)

PFS 75 Quartile (months)

PFS 95% Confidence Interval (months)

6.90

17.45

19.57

(3.00, 23.69)

21.441 (p-value = 0.01488)

Astrocytoma-mutant II

37.20

43.20

55.63

(35.7, 60.0)

Oligodendroglioma II

47.42

59.20

88.28

(46.3, 91.2)

16.6 (95% CI: 1.89-38.8), p = 0.02429

Figure 1: PRISMA Flow Diagram Records identified through database searching MEDLINE: 3760 EMBASE: 4367 CENTRAL: 184 (n = 8311)

Additional records identified through other sources (n = 4)

Records after duplicates removed (n = 7542)

Records screened by abstract and title for potential eligibility (n =7542) Full-text articles assessed for eligibility (n = 67)

Studies included in qualitative synthesis (n = 19)

Studies included in quantitative synthesis (n = 16)

Records excluded (n = 7475)

Full-text articles excluded, with reasons (did not provide adequate data set; did not meet inclusion criteria; repeated data as prior paper/conference abstract) (n = 48)

Abbreviations OS: Overall Survival PFS: Progression Free Survival PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analysis LGG: Low Grade Gliomas RCT: Randomized Controlled Trials CNS: Central Nervous System IDH: isocitrate dehydrogenase CT: chemotherapy RT: radiotherapy MeSH: Medical Subheadings CENTRAL: Cochrane Central Register of Controlled Trails IoN: Institute of Neurology A-mt II: IDH-mt non-codeletion astrocytoma OD-II: IDH-mt codeletion oligodendroglioma A-wt II: IDH-wt astrocytoma