Pattern Of Failure In Idh Mutated, Low Grade Glioma After Radiotherapy - Implications For Margin Reduction

•The pattern of failure for IDH mutated, grade 2 glioma is predominantly in – field.•Pattern of failure in this study is comparable to historic data.•A lower percentage of recurrences are covered by a 5 mm expansion of the GTV. Low grade glioma (LGG) is a group of relatively slow growing primary brain neoplasms, chiefly occurring between 30 and 50 years of age [[1]Forst D.A. Nahed B.V. Loeffler J.S. Batchelor T.T. Low-grade gliomas.Oncologist. 2014; 19: 403-413Crossref PubMed Scopus (93) Google Scholar]. With recent advances in molecular genetics, it has been found that molecular subtype is a better predictor of prognosis than classical histology [2Eckel-Passow J.E. Lachance D.H. Molinaro A.M. Walsh K.M. Decker P.A. Sicotte H. et al.Glioma groups based on 1p/19q, IDH, and TERT promoter mutations in tumors.N Engl J Med. 2015; 372: 2499-2508Crossref PubMed Scopus (1001) Google Scholar, 3Gravendeel L.A.M. Kouwenhoven M.C.M. Gevaert O. de Rooi J.J. Stubbs A.P. Duijm J.E. et al.Intrinsic gene expression profiles of gliomas are a better predictor of survival than histology.Cancer Res. 2009; 69: 9065Crossref PubMed Scopus (403) Google Scholar]. The 2016 WHO classification of glioma is based on a genotype-driven classification of diffuse gliomas [[4]Wesseling P. Capper D.W.H.O. Classification of gliomas.Neuropathol Appl Neurobiol. 2016; 2018: 139-150Google Scholar]. The grade 2 gliomas have been subdivided along the presence or absence of a mutation in the isocytrate dehydrogenase 1 or 2 (IDH). The group of tumors with a mutation present in the IDH gene (IDHmG) have a relatively favorable prognosis, while group of IDH wildtype (IDHwt) tumors have a prognosis more akin to glioblastoma [5Hartmann C. Hentschel B. Wick W. Capper D. Felsberg J. Simon M. et al.Patients with IDH1 wild type anaplastic astrocytomas exhibit worse prognosis than IDH1-mutated glioblastomas, and IDH1 mutation status accounts for the unfavorable prognostic effect of higher age: implications for classification of gliomas.Acta Neuropathol. 2010; 120: 707-718Crossref PubMed Scopus (534) Google Scholar, 6Tesileanu C.M.S. Dirven L. Wijnenga M.M.J. Koekkoek J.A.F. Vincent A.J.P.E. Dubbink H.J. et al.Survival of diffuse astrocytic glioma, IDH1/2 wildtype, with molecular features of glioblastoma, WHO grade IV: a confirmation of the cIMPACT-NOW criteria.Neuro-Oncol. 2019; 22: 515-523Crossref Scopus (31) Google Scholar]. The objective of radiotherapy in low grade glioma is an extended period of local control. The place of postoperative radiotherapy and chemotherapy in grade 2 glioma was established by the results of several multicenter trials [7Buckner J.C. Shaw E.G. Pugh S.L. Chakravarti A. Gilbert M.R. Barger G.R. et al.Radiation plus procarbazine, CCNU, and vincristine in low-grade glioma.N Engl J Med. 2016; 374: 1344-1355Crossref PubMed Scopus (494) Google Scholar, 8Fisher B.J. Hu C. Macdonald D.R. Lesser G.J. Coons S.W. Brachman D.G. et al.Phase 2 study of temozolomide-based chemoradiation therapy for high-risk low-grade gliomas: preliminary results of radiation therapy oncology group 0424.Int J Radiat Oncol Biol Phys. 2015; 91: 497-504Abstract Full Text Full Text PDF PubMed Scopus (103) Google Scholar, 9Shaw E.G. Wang M. Coons S.W. Brachman D.G. Buckner J.C. Stelzer K.J. et al.Randomized trial of radiation therapy plus procarbazine, lomustine, and vincristine chemotherapy for supratentorial adult low-grade glioma: initial results of RTOG 9802.J Clin Oncol. 2012; 30: 3065-3070Crossref PubMed Scopus (217) Google Scholar]. After a disease-free interval, a subset of low grade glioma are known to undergo malignant transformation, almost invariably inside or in close proximity to the radiation field. Improvements in imaging, neurosurgical technique, and the introduction of adjuvant chemotherapy over the past 20 years have increased the prognosis of LGG patients considerably. As such, there has been a shift in focus from achieving disease control towards reducing the late adverse effects of radiotherapy. The use of radiotherapy, especially the large fields applied in the past, has been implicated in the onset of late neurocognitive decline [[10]Douw L. Klein M. Fagel S.S. van den Heuvel J. Taphoorn M.J. Aaronson N.K. et al.Cognitive and radiological effects of radiotherapy in patients with low-grade glioma: long-term follow-up.Lancet Neurol. 2009; 8: 810-818Abstract Full Text Full Text PDF PubMed Scopus (452) Google Scholar]. Recent trials have reduced the GTV-CTV expansion to 10 mm (NRG BN005, NCT03180502) or 5 mm (EORTC 1635, NCT03763422). The current working document of the Dutch Platform for Radiotherapy in Neuro-Oncology advises a margin of 5 mm to be used in clinical care. However, effect of these smaller fields on pattern of failure is not yet known. We sought to assess the safety of a CTV margin reduction to 5 mm using a retrospective analysis of historical treatments of IDHmt low grade glioma patients using the 2011 RANO criteria for progressive disease [[11]van den Bent M.J. Wefel J.S. Schiff D. Taphoorn M.J. Jaeckle K. Junck L. et al.Response assessment in neuro-oncology (a report of the RANO group): assessment of outcome in trials of diffuse low-grade gliomas.LancetOncol. 2011; 12: 583-593Scopus (333) Google Scholar]. We reviewed the charts of all patients treated with radiotherapy for histologically confirmed low grade glioma between 1-1-2007 and 31-12-2017 in Erasmus MC. Of the patients exhibiting disease progression, the original planning CT, structure set, and dose object were retrieved. In patients in which the IDH status was not known, IDH was sequenced from archived material. Finally, a number of cases in which IDH status was known from a separate project [[12]Jaspers J. Mèndez Romero A. Hoogeman M.S. van den Bent M. Wiggenraad R.G.J. Taphoorn M.J.B. et al.Evaluation of the hippocampal normal tissue complication model in a prospective cohort of low grade glioma patients—An analysis within the EORTC 22033 clinical trial.Front Oncol. 2019; 9Crossref PubMed Scopus (10) Google Scholar] were found to have disease progression on follow-up. Data from these cases was requested from their treating centers. The study was conducted according to the principles of the Declaration of Helsinki (59th WMA General Assembly, Seoul, October 2008) and in accordance with the local medical research regulations. The study protocol was presented to the local Medical Ethics Committee (MEC-2019-255) and considered not subject to the Medical Research Involving Human Subjects Act. Resection status was defined as either biopsy only, complete resection (if no residual tumor mass was reported on postoperative imaging) or partial (if residual tumor mass was present). Follow up MRI’s typically included of at least T1 weighed pre- and post-contrast, 2d T2 weighed and FLAIR images. All available MRI’s were reviewed. Type of recurrence was defined according to Response Assessment in Neuro Oncology (RANO) criteria as either enhancing (development of a new contrast enhancing lesion) or non-enhancing (an increase of 25% in perpendicular diameter of T2 abnormalities without enhancement). The date of progressive disease (PDRANO) was the date of the first MRI that fulfilled the RANO criteria for recurrence. Time to progression (TTP) was defined as the interval between the last RT fraction and the date of progressive disease according to the RANO criteria. In all centers, it was customary to confirm the diagnosis of recurrence in a multidisciplinary neuro-oncological tumor board before a next intervention was started. The date on which the diagnosis of recurrence was confirmed by the tumor board was defined as “tumor-board progressive disease” (PDtb). In order to avoid transient contrast-enhancing lesions being interpreted as disease recurrence we confirm the presence of these lesions over sequential MRIs between PDRANO until PDtb. Adjuvant chemotherapy was defined as chemotherapy started within 3 months after completion of radiotherapy in absence of disease progression. The MRI at time of PDRANO was rigidly matched to the original planning CT using MIM (MIM software, version 6.3.9, Cleveland, Ohio). In patients with an enhancing recurrence, the recurrence volume (rTV) was defined as the area of pathological enhancement on T1 series. In patients without an enhancing recurrence, the recurrence volume was defined as the areas of T2 hypo-attenuation that exhibited progression over the preceding 12 months. The volumes were delineated by AMR and JJ (radiation oncologists), and delineations were approved by MvdB (neurologist). A hypothetical CTV of 5 mm (CTV5mm) was generated by creating a 5 mm expansion of the original GTV, and limiting this to within the original CTV. The overlap between the rTV, the original CTV, the CTV5mm and the original 95% isodose volume was calculated. Recurrence was classified as either central (>95%), inside (>80–95%), edge (>20–80%), or outside (≤20%) of the original CTV. Dose volume histograms (DVH) were generated for all recurrences. The distribution of recurrences with regards to the original CTV and the CTV5mm was compared using a two-way ANOVA. Overall survival and progression free survival were assessed using a Kaplan-Meier analysis. Statistical analysis was done in R (www.r-project.org) and SPSS (IBM Corp., IBM SPSS Statistics for Windows, Version 25.0.0.1, Armonk, New York). Between 1-5-2007 and 31-12-2017, 113 patients underwent radiotherapy for low grade glioma. A recurrence was diagnosed in 56 patients. Radiotherapy planning and delineation could be retrieved in 49 of these patients. In 35 of these patients a positive IDH mutation status was found. Four additional fully documented cases with known IDH status and disease recurrence were added from two centers. The final dataset comprised 39 IDHmG patients with known recurrence. Patient characteristics are summarized in Table 1.Table 1Patient characteristics.Age (years)42.1(95% CI 39.1 –45.1)SexMale2666.6%Female1333.3%HemisphereRight1641.0%Left1948.7%Both410,3%LobeFrontal2256.4%Temporal512.8%Parietal410.3%Occipital410.3%Brainstem12.6%Overlapping lesion37.7%Resection statusBiopsy1128.2%Partial or subtotal resection2359.0%Gross total resection410.3%Unknown12.6%1p/19q codeletionPresent1743.6%Absent1743.6%Undetermined512.8%Technique3DCT2564.1%IMRT1435.9%CTV margin10 mm1025.6%15 mm2974.4%CTV volume (cc)294(95% CI 252–336)Adjuvant chemotherapyNone3589.7%Temozolomide25.1%PCV25.1% Open table in a new tab Mean age at diagnosis was 42.1 years (95% CI 39.6–45.7). Resection status was gross total resection in 4 patients (10%), partial resection in 23 (59%), biopsy in 11 patients (28%), and unknown in one patient (3%). A 1p/19q deletion was present in 17 patients (44%), absent in 17 patients (44%) and undetermined in 5 patients (13%). Median interval between surgery and start of radiotherapy was 0.3 years (range 0.2–12.0). All patients were treated to a dose of 50.4 Gy in 28 fractions (ICRU 50). Patients were treated with either 3DCT (64%) or IMRT (36%). GTV-CTV margin was 15 mm in 29 patients (74%), and 10 mm in 10 patients (26%). The mean CTV volume was 294 cc (95% CI 252–336). PTV margin was 5 mm in all patients. Four patients (10%) were treated with adjuvant chemotherapy. By the time of analysis the median duration of follow-up from end of radiotherapy was 5.0 years (range 1.4–11.4). Median overall survival was 5.6 years (range 1.3–11.4), with 24 patients having died of disease. Median TTP was 2.8 years (range 0.6–9.3). In 21 patients, the date of PDRANO was within 14 days of the date of PDtb. However, in 18 patients the date of PDRANO predated the date PDtb by a median of 0.5 years (range 0.1–3.0). Time intervals and survival are summarized in Fig. 1. At the time of PDRANO 34 patients developed an enhancing recurrence (87%) and five patients (13%) developed a non-enhancing recurrence. The mean volume of the rTV of enhancing recurrences was 5.2 cc (range 0.1–37.7). The mean volume of the rTV of non-enhancing recurrences was 32.8 cc (range 2.4–140.3). With regards to the original CTV, recurrences were classified as central in 32 (82%), inside in 3 (8%), edge in 1 (2%) and outside in 3 patients (8%). Almost all recurrences (92%) were covered by the 95% isodose line, three (8%) were out-field. Based on the hypothetical CTV5mm, recurrences would have been central in 26 (66%), inside in 4 (10%), edge in 5 (13%), and outside in 4 (10%) patients (Fig. 2). The difference in distribution of recurrence clas was significant (p = 0.005). See Fig. 3 for two examples of recurrences. Owing to the low number of non-enhancing recurrences, we were unable to test whether the distribution of recurrence class differs between enhancing and non-enhancing recurrences. However, the probability of a central recurrence was higher in enhancing recurrences (88%) than in non-enhancing recurrences (40%, p = 0.03, see Supplementary data 1). The mean dose to 98% of the rTV (D98%) was 50.4 Gy in the recurrences classified as central with respect to the CTV (range 48.4–54.4), 48.8 Gy in the inside category (range 48.6–49.2), 44.1 Gy in the one edge recurrence, and 14.7 Gy (range 1.5–34.4) in the outside category (see Supplementary data 2 and 3).Fig. 3Two examples of recurrences. The dose distribution and the structure set are superimposed on the MRI at the time of recurrence. The GTV is in yellow, the original CTV is dark blue, the PTV is red. The CTV5mm is in light blue. The rTV is in pink. To the left is a T1 recurrence classified as “central” with regards to the original CTV, “inside” with regards to the CTV5mm, and “in field” with regards to the 95% isodose. To the right is a parahippocampal (and infratentorial) T2 recurrence classified as out-field.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The size of the treatment field has been a point of contention since the introduction of radiotherapy for glioma. Historically, proponents of partial brain techniques argued a smaller treatment field would lead to less adverse effects and potential for dose escalation [13Todd D.H. Choice of Volume in the X-ray treatment of supratentorial gliomas.Br J Radiol. 1963; 36: 645-649Crossref PubMed Scopus (16) Google Scholar, 14Leibel S.A. Sheline G.E. Wara W.M. Boldrey E.B. Nielsen S.L. The role of radiation therapy in the treatment of astrocytomas.Cancer. 1975; 35: 1551-1557Crossref PubMed Scopus (151) Google Scholar, 15Scanlon P.W. Taylor W.F. Radiotherapy of Intracranial Astrocytomas: Analysis of 417 Cases Treated from 1960 through 1969.Neurosurgery. 1979; 5: 301-308Crossref PubMed Scopus (110) Google Scholar], while proponents of whole brain radiotherapy emphasized the risk of out-field failure in light of uncertainties in target localization [16Fazekas J.T. Treatment of grades I and II brain astrocytomas. the role of radiotherapy.Int J Radiat Oncol Biol Phys. 1977; 2: 661-666Abstract Full Text PDF PubMed Scopus (114) Google Scholar, 17Salazar O.M. Rubin P. McDonald J.V. Feldstein M.L. Patterns of failure in intracranial astrocytomas after irradiation: analysis of dose and field factors.Am J Roentgenol. 1976; 126: 279-292Crossref PubMed Scopus (55) Google Scholar]. Following the availability of CT and MRI imaging, the landmark trials in low grade glioma of the 1990s and 2000s all adopted target volumes based on some form of a margin around a lesion visible on imaging of 15 to 20 mm (Table 2). As no prospective data on smaller treatment margins exists, a CTV margin of 10–15 mm is still standard of care in many centers.Table 2Overview of specified treatment margins in selected trials, and published pattern of failure for low grade glioma. Note that ICRU29 definition defined a target volume, and ICRU50 report and on define a CTV and a PTV. See [36]Purdy J.A. Current ICRU definitions of volumes: limitations and future directions.Semin Radiat Oncol. 2004; 14: 27-40Crossref PubMed Scopus (77) Google Scholar for an illustrated overview.ProcedureTargetICRU definitionCompleted trialsEORTC 22,844 [31]Karim A.B. Maat B. Hatlevoll R. Menten J. Rutten E.H. Thomas D.G. et al.A randomized trial on dose-response in radiation therapy of low-grade cerebral glioma: European Organization for Research and Treatment of Cancer (EORTC) Study 22844.Int J Radiat Oncol Biol Phys. 1996; 36: 549-556Abstract Full Text PDF PubMed Scopus (541) Google ScholarCT enhancing lesion + 20 mmCT edema + 10 mmTarget volumeICRU29EORTC 22,845 [37]van den Bent M.J. Afra D. de Witte O. Ben Hassel M. Schraub S. Hoang-Xuan K. et al.Long-term efficacy of early versus delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults: the EORTC 22845 randomised trial.Lancet. 2005; 366: 985-990Abstract Full Text Full Text PDF PubMed Scopus (685) Google ScholarMRI T2 abnormalities + 20 mmTarget volumeICRU29RTOG 9802 [9]Shaw E.G. Wang M. Coons S.W. Brachman D.G. Buckner J.C. Stelzer K.J. et al.Randomized trial of radiation therapy plus procarbazine, lomustine, and vincristine chemotherapy for supratentorial adult low-grade glioma: initial results of RTOG 9802.J Clin Oncol. 2012; 30: 3065-3070Crossref PubMed Scopus (217) Google ScholarMRI T2 abnormalities + 20 mmField edgeICRU29Intergroup [32]Shaw E. Arusell R. Scheithauer B. O'Fallon J. O'Neill B. Dinapoli R. et al.Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a North Central Cancer Treatment Group/Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group study.J Clin Oncol. 2002; 20: 2267-2276Crossref PubMed Scopus (547) Google ScholarLesion on CT or MRI + 20 mmTarget volumeICRU29EORTC 22,033–26,033 [20]Baumert B.G. Hegi M.E. van den Bent M.J. von Deimling A. Gorlia T. Hoang-Xuan K. et al.Temozolomide chemotherapy versus radiotherapy in high-risk low-grade glioma (EORTC 22033–26033): a randomised, open-label, phase 3 intergroup study.LancetOncol. 2016; 17: 1521-1532Scopus (232) Google ScholarMRI T1 enhancement and T2 abnormalities + 15 mmCTVICRU50Ongoing trialsNRG BN005NCT0318050210 mmCTVICRU50EORTC 1635NCT03763422 (QA guideline)5 mmCTVICRU50 Open table in a new tab To our knowledge, this study is the first to use a volumetric approach to classify the pattern of recurrence in IDH mutated, low grade glioma as defined by the new 2016 WHO classification. Although varying in methodology, for example, a centroid approach [[18]Jakola A.S. Bouget D. Reinertsen I. Skjulsvik A.J. Sagberg L.M. Bo H.K. et al.Spatial distribution of malignant transformation in patients with low-grade glioma.J Neurooncol. 2020; 146: 373-380Crossref PubMed Scopus (2) Google Scholar] or visual methods [[19]Kamran S.C. Dworkin M. Niemierko A. Bussiere M. Oh K.S. Loeffler J.S. et al.Patterns of failure among patients with low-grade glioma treated with proton radiation therapy.Pract Radiat Oncol. 2019; 9: e356-e361Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar], all known studies investigating pattern of recurrence find the vast majority of failures to occur within high-dose area of the original treatment field (see Table 3). In this study, we find a similar pattern of failure in patients treated between 2007 and 2017 with the dose of 50.4 Gy in 28 fractions regarded as standard, using a GTV-CTV expansion, a PTV margin, and photon therapy planning techniques (3DCT, IMRT) that represented standard of care. The results from the proton therapy cohort presented by Kamran [[19]Kamran S.C. Dworkin M. Niemierko A. Bussiere M. Oh K.S. Loeffler J.S. et al.Patterns of failure among patients with low-grade glioma treated with proton radiation therapy.Pract Radiat Oncol. 2019; 9: e356-e361Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar], which dates from 2005 to 2015, suggest the pattern of failure in proton beam therapy is comparable.Table 3Overview of published studies in low grade glioma with pattern of failure data.MarginNumber of recurrencesIn fieldField edgeOut of fieldPu, 1994 [38]Pu A.T. Sandler H.M. Radany E.H. Blaivas M. Page M.A. Greenberg H.S. et al.Low grade gliomas: preliminary analysis of failure patterns among patients treated using 3D conformal external beam irradiation.Int J Radiat Oncol Biol Phys. 1995; 31: 461-466Abstract Full Text PDF PubMed Scopus (45) Google Scholar10–30 mm to target volume11100%0%0%Rudoler, 1998 [39]Rudoler S. Corn B.W. Werner-Wasik M. Flanders A. Preston P.E. Tupchong L. et al.Patterns of tumor progression after radiotherapy for low-grade gliomas: analysis from the computed tomography/magnetic resonance imaging era.Am J Clin Oncol. 1998; 21: 23-27Crossref PubMed Scopus (34) Google Scholar*The study population included 8 cases treated with whole brain radiotherapy.20 mm to target volume16100%0%0%van den Bent, 2005 [37]van den Bent M.J. Afra D. de Witte O. Ben Hassel M. Schraub S. Hoang-Xuan K. et al.Long-term efficacy of early versus delayed radiotherapy for low-grade astrocytoma and oligodendroglioma in adults: the EORTC 22845 randomised trial.Lancet. 2005; 366: 985-990Abstract Full Text Full Text PDF PubMed Scopus (685) Google Scholar20 mm to target volume9490%5%4%Shaw, 2002 [32]Shaw E. Arusell R. Scheithauer B. O'Fallon J. O'Neill B. Dinapoli R. et al.Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a North Central Cancer Treatment Group/Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group study.J Clin Oncol. 2002; 20: 2267-2276Crossref PubMed Scopus (547) Google Scholar20 mm to target volume6592%3%5%Kamran, 2019 [19]Kamran S.C. Dworkin M. Niemierko A. Bussiere M. Oh K.S. Loeffler J.S. et al.Patterns of failure among patients with low-grade glioma treated with proton radiation therapy.Pract Radiat Oncol. 2019; 9: e356-e361Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar7–15 mm to CTV4176%12%12%This study10–15 mm to CTV3992%0%8%* The study population included 8 cases treated with whole brain radiotherapy. Open table in a new tab The cases in our cohort were selected for disease progression, resulting in a median TTP of only 2.8 years after radiotherapy at a median duration of follow-up of 5.0 years. Contrasting this, the median TTP in the entire radiotherapy – only group of EORTC 22033–26033 [[20]Baumert B.G. Hegi M.E. van den Bent M.J. von Deimling A. Gorlia T. Hoang-Xuan K. et al.Temozolomide chemotherapy versus radiotherapy in high-risk low-grade glioma (EORTC 22033–26033): a randomised, open-label, phase 3 intergroup study.LancetOncol. 2016; 17: 1521-1532Scopus (232) Google Scholar] was 3.8 years at a median duration of follow – up of 4.0 years, reflecting the case selection in this study. It is notable that all but 4 patients treated in this study date from before the introduction of adjuvant chemotherapy. Both PCV and temozolomide chemotherapy are known to inhibit tumor growth [21Taal W. van der Rijt C.C.D. Dinjens W.N.M. Sillevis Smitt P.A.E. Wertenbroek A.A.A.C.M. Bromberg J.E.C. et al.Treatment of large low-grade oligodendroglial tumors with upfront procarbazine, lomustine, and vincristine chemotherapy with long follow-up: a retrospective cohort study with growth kinetics.J Neurooncol. 2015; 121: 365-372Crossref PubMed Scopus (22) Google Scholar, 22Izquierdo C. Alentorn A. Idbaih A. Simó M. Kaloshi G. Ricard D. et al.Long-term impact of temozolomide on 1p/19q-codeleted low-grade glioma growth kinetics.J Neurooncol. 2018; 136: 533-539Crossref PubMed Scopus (13) Google Scholar], and the use of chemotherapy is associated with a benefit in PFS in IDHmG [7Buckner J.C. Shaw E.G. Pugh S.L. Chakravarti A. Gilbert M.R. Barger G.R. et al.Radiation plus procarbazine, CCNU, and vincristine in low-grade glioma.N Engl J Med. 2016; 374: 1344-1355Crossref PubMed Scopus (494) Google Scholar, 9Shaw E.G. Wang M. Coons S.W. Brachman D.G. Buckner J.C. Stelzer K.J. et al.Randomized trial of radiation therapy plus procarbazine, lomustine, and vincristine chemotherapy for supratentorial adult low-grade glioma: initial results of RTOG 9802.J Clin Oncol. 2012; 30: 3065-3070Crossref PubMed Scopus (217) Google Scholar]. As the number of patients that received adjuvant chemotherapy in this study is low, the influence of adjuvant chemotherapy on the pattern of failure is not known. The definition of disease progression may influence the pattern of failure. As the recurrence volume will increase over time, mostly in a centripetal manner, definition of recurrence at a later time point will lead to an increasing number of failures in the “inside” or “edge” categories. As an additional observation, we found that when retrospectively assessing all imaging, PDRANO was found to predate PDtb almost half of all patients. A similar finding appears in Izquierdo et al [[22]Izquierdo C. Alentorn A. Idbaih A. Simó M. Kaloshi G. Ricard D. et al.Long-term impact of temozolomide on 1p/19q-codeleted low-grade glioma growth kinetics.J Neurooncol. 2018; 136: 533-539Crossref PubMed Scopus (13) Google Scholar], reporting an interval between retrospectively assessed RANO – progressive disease and the next intervention of 11 months. New contrast - enhancing lesions after radiotherapy are not uncommon and some represent benign post-treatment changes [23de Wit M.C. de Bruin H.G. Eijkenboom W. Sillevis Smitt P.A. van den Bent M.J. Immediate post-radiotherapy changes in malignant glioma can mimic tumor progression.Neurology. 2004; 63: 535-537Crossref PubMed Scopus (278) Google Scholar, 24van West S.E. de Bruin H.G. van de Langerijt B. Swaak-Kragten A.T. van den Bent M.J. Taal W. Incidence of pseudoprogression in low-grade gliomas treated with radiotherapy.Neuro Oncol. 2017; 19: 719-725PubMed Google Scholar]. It is likely that new contrast – enhancing lesion are observed for a time period in follow up, before they are considered indicative of disease recurrence. In this study, the contrast – enhancing lesions that were delineated at PDRANO were identified with the benefit of hindsight. As such, all lesions delineated developed into the actual recurrence confirmed by the tumor board at PDtb. There are several factors other than the GTV-CTV and tumor size that determine the size of the treatment field. The size of the GTV is determined in part by the choice of neurosurgical resection, as the resection cavity will be included in addition to residual tumor volume. Larger extent resections have been shown to influence progression-free survival in some series [25Wijnenga M.M.J. French P.J. Dubbink H.J. Dinjens W.N.M. Atmodimedjo P.N. Kros J.M. et al.The impact of surgery in molecularly defined low-grade glioma: an integrated clinical, radiological, and molecular analysis.Neuro-Oncol. 2017; 20: 103-112Crossref Scopus (107) Google Scholar, 26Patel T. Bander E.D. Venn R.A. Powell T. Cederquist G.Y. Schaefer P.M. et al.The role of extent of resection in IDH1 wild-type or mutant low-grade gliomas.Neurosurgery. 2018; 82: 808-814Crossref PubMed Scopus (25) Google Scholar]. The use of additional imaging modalities, such as PET, may also increase the size of the GTV [[27]Albert N.L. Weller M. Suchorska B. Galldiks N. Soffietti R. Kim M.M. et al.Response assessment in neuro-oncology working group and European association for neuro-oncology recommendations for the clinical use of PET imaging in gliomas.Neuro-oncology. 2016; 18: 1199-1208Crossref PubMed Scopus (305) Google Scholar]. Even with the use of modern imaging, the inter – observer variability in GTV delineation is known to be substantial in diffuse glioma [[28]Abrunhosa-Branquinho A.N. Bar-Deroma R. Collette S. Clementel E. Liu Y. Hurkmans C.W. et al.Radiotherapy quality assurance for the RTOG 0834/EORTC 26053–22054/NCIC CTG CEC.1/CATNON intergroup trial “concurrent and adjuvant temozolomide chemotherapy in newly diagnosed non-1p/19q deleted anaplastic glioma”: Individual case review analysis.Radiother Oncol. 2018; 127: 292-298Abstract Full Text Full Text PDF PubMed Scopus (8) Google Scholar]. Uncertainty in target identification is normally incorporated in the PTV margin, along with other random and systemic errors in planning and dose delivery, and as such may influence the size of the treatment field [29van Herk M. Errors and margins in radiotherapy.Semin Radiat Oncol. 2004; 14: 52-64Crossref PubMed Scopus (950) Google Scholar, 30Prescribing, Recording, and Reporting Photon-Beam Intensity-Modulated Radiation Therapy (IMRT): Contents Journal of the International Commission on Radiation Units and Measurements. 2016;10:NP-NP.Google Scholar]. In interpreting pattern of failure, in-field failure occurs when insufficient dose was delivered to kill all tumor cells inside the CTV. Contrasting this, edge failure might be interpreted as a result of a geographic miss, occurring when the chosen treatment field failed to encompass the future site of relapse. In low-grade glioma it is known that dose escalation will result in equal survival at best, with a potentially negative impact on quality of life [31Karim A.B. Maat B. Hatlevoll R. Menten J. Rutten E.H. Thomas D.G. et al.A randomized trial on dose-response in radiation therapy of low-grade cerebral glioma: European Organization for Research and Treatment of Cancer (EORTC) Study 22844.Int J Radiat Oncol Biol Phys. 1996; 36: 549-556Abstract Full Text PDF PubMed Scopus (541) Google Scholar, 32Shaw E. Arusell R. Scheithauer B. O'Fallon J. O'Neill B. Dinapoli R. et al.Prospective randomized trial of low- versus high-dose radiation therapy in adults with supratentorial low-grade glioma: initial report of a North Central Cancer Treatment Group/Radiation Therapy Oncology Group/Eastern Cooperative Oncology Group study.J Clin Oncol. 2002; 20: 2267-2276Crossref PubMed Scopus (547) Google Scholar]. Since an increase in dose using current treatment fields is not opportune, and in light of the improving prognosis of IDHmG patients, it would be interesting to reduce field size while maintaining the current pattern of failure. Future approaches may include individualized treatment margins based on molecular criteria that define high – and low risk groups [[33]Geurts M. van den Bent M.J. On high-risk, low-grade glioma: What distinguishes high from low?.Cancer. 2019; 125: 174-176Crossref PubMed Scopus (5) Google Scholar]. Additionally, incorporating a data driven approach based on recurrence probabilities [[34]Peeken J.C. Molina-Romero M. Diehl C. Menze B.H. Straube C. Meyer B. et al.Deep learning derived tumor infiltration maps for personalized target definition in Glioblastoma radiotherapy.Radiother Oncol. 2019; 138: 166-172Abstract Full Text Full Text PDF PubMed Scopus (13) Google Scholar], or an imaging-derived approach based on models of tumor spread [[35]Trip A.K. Jensen M.B. Kallehauge J.F. Lukacova S. Individualizing the radiotherapy target volume for glioblastoma using DTI-MRI: a phase 0 study on coverage of recurrences.Acta Oncol. 2019; 58: 1532-1535Crossref PubMed Scopus (4) Google Scholar] may lead to smaller fields with an identical pattern of recurrence. Such an approach would, however, require confirmation in prospective studies with long follow-up. This study has some limitations, mostly stemming from its retrospective design. It is important to note that the observation of a lower percentage of recurrence volume covered by a retrospectively constructed 5 mm CTV margin should not be interpreted as evidence that smaller margins will lead to increased edge relapse. It can be concluded, however, that not all recurrence volumes are within 5 mm of the GTV. Since no prospective data on treatment margins below 10–15 mm exist in IDHmG, we feel GTV-CTV expansions below 10 mm are to be used cautiously. JJ wrote the manuscript and conducted the statistical analysis. Volumes were delineated by AMR and JJ, and approved by MvdB. AMR, RN and MvdB supervised the project. All authors contributed to the final version of the manuscript. An abstract of this work was presented as a poster at ASTRO 2020. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. The following are the Supplementary data to this article: Download .docx (.09 MB) Help with docx files Supplementary data 1