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