Abstract
Background
The Response Assessment in Pediatric Neuro-Oncology (RAPNO) working group for diffuse intrinsic pontine glioma (DIPG) recently published its recommendations. We aim to test the operative performance of the RAPNO DIPG criteria imaging component by retrospectively applying it to a patient sample from the International DIPG/DMG Registry (IDIPGR).
Methods
Longitudinal MRIs for 46 patients were independently reviewed by 2 pediatric neuro-radiologists. Utilizing RAPNO DIPG imaging criteria for the pontine lesions, response was categorized as complete response (CR), partial response (PR), stable disease (SD), or progressive disease (PD). The response category for each MRI was compared between 2 readers and classified as concordant if they agreed, minor discordant if one reported SD and the other reported PR or PD, and major discordant if one reported PR and the other reported PD.
Results
A total of 277 paired MRIs were analyzed, and 124 paired MRIs were evaluated for concordance. The response category was concordant between readers in 84 (68%) MRI comparisons. In 31 MRI comparisons (25%) the reads were minor discordant, and major discordant in 9 (7%). No CRs were reported. Minor discordant cases were within 10% of the boundary for PR or PD in 20 (65%) of these cases. The median difference between the 2 readers’ measurements was 2 mm (range 0-29 mm).
Conclusion
This study demonstrated that RAPNO DIPG imaging criteria can be applied with concordance or minor discordance between readers in 93% of the cases. Discordant measurements were largely at the boundaries of response type.
Keywords: DIPG, DMG, IDIPG/DMG registry, Response assessment in pediatric neuro-oncology (RAPNO), response evaluation
Key Points.
1 Using RAPNO DIPG criteria, response categories between reviewers were concordant/ minor discordant in 93% cases
2 When discordant, most of the disagreements were at the margins of the response categories
Importance of the Study.
Disease-specific RAPNO criteria were developed to standardize the response assessment in the clinical trials. But, RAPNO criteria have implications beyond clinical trials, and can also be applied to standardize response assessment in the clinical setting. The RAPNO DIPG criteria have not been tested retrospectively or prospectively, and the current study aims to test the operative performance of the RAPNO DIPG criteria in the decentralized setting. Two pediatric neuro-radiologists from different institutions independently reviewed MRIs of a sample of patients from the IDIPG/DMG registry using RAPNO DIPG criteria for pontine lesions, and their response categories were compared. Our study showed that RAPNO DIPG criteria for pontine lesions can be applied with concordance/minor discordance in almost all cases. We also discuss the challenges faced by neuro-radiologists using RAPNO DIPG criteria for image review.
Historically, diffuse intrinsic pontine glioma (DIPG) has been diagnosed based on the clinical presentation and imaging features. Biopsy is increasingly being used to understand molecular features but is not required for the diagnosis of DIPG. DIPG tissue commonly harbors H3K27 mutation, and the World Health Organization (WHO) CNS tumor classification 2021 defines diffuse midline glioma based on this well-described molecular change.1–5 DIPG has a dismal outcome, with a median survival of 11 months and a 2-year overall survival of <10 %.6 Given the poor survival and absence of a defined standard of care, patients with DIPG are encouraged to enroll in clinical trials.7–16 Reporting a clinical trial result and comparing it to other trials requires objective response evaluation criteria. Clinical trials historically used RECIST, RANO, or WHO criteria for response evaluation, but these older criteria do not account for the multiple complexities of modern DIPG treatments like pseudo-progression, focal drug delivery, steroid, and anti-angiogenic agents use.8 To develop disease-specific response assessment guidelines for pediatric CNS tumors, Response Assessment in Pediatric Neuro-Oncology (RAPNO) working groups were established consisting of an international panel of pediatric and adult neuro-oncologists, neuro-radiologists, radiation oncologists, neurosurgeons, and other relevant sub-specialists.17 Working groups published tumor-specific response assessment guidelines between 2018 and 2023.18–23 To test the operative performance of the imaging component of RAPNO DIPG criteria, we retrospectively applied these recommendations to MRI images housed in the International Diffuse Intrinsic Pontine Glioma/Diffuse midline glioma registry (IDIPGR) to determine the concordance between 2 pediatric neuro-radiologists.
Methods
The IDIPGR is housed at Cincinnati Children’s Hospital Medical Center and patient referral to the registry has been previously described.24 Referred patients are consented by registry staff and physicians. This study was approved by the International DIPG/DMG registry scientific review board and the Central Michigan University Investigational Review Board (IRB). This study was performed on a preselected sample of 312 MRIs from 46 patients and the selection criteria have been described previously.25 Briefly, selection criteria included: baseline MRI findings supportive of a DIPG based on the central review; the patient must have received upfront RT; and a baseline MRI and at least 2 post-RT MRIs that met quality control standards. Biopsy of the pontine lesion was not required, but when available, it needed to be consistent with infiltrating grade II-IV gliomas (diffuse midline glioma, H3K27M-mutant, glioblastoma, anaplastic astrocytoma, diffuse astrocytoma). Patients with the majority of tumors in the cerebellum or extensive infiltrative tumors in the cerebellum were excluded. Patients with sub-ventricular tumor spread or noncontiguous metastatic disease denoting progression in the absence of progression in the primary disease were excluded. MRIs were reviewed by 2 pediatric neuro-radiologists (MK and MG) independently using RAPNO DIPG criteria. Both have more than 5 years of post-fellowship experience. They were educated about RAPNO DIPG criteria and had the RAPNO DIPG article available for reference.18 An Ambrahealth website account was created through which both radiologists had access to preselected MRIs. Ambrahealth (ambrahealth.com) is a cloud-based platform to store and share medical images. Both radiologists reviewed test MRIs in a virtual meeting for training purposes; test MRIs were not included in the study. Readers were blinded to clinical information, institutional reports, and measurements from the other reader. The measurements were performed on a dedicated radiology workstation.
There was some variability in imaging sequences and protocols between the treating institutions. All MRIs evaluated for this study had axial T2-weighted image slice thickness between 3–5 mm and 0–1 mm inter-slice gap.25 Imaging sequences used for this study were T1-weighted (pre and postcontrast), and T2-weighted or FLAIR sequences to record long-axis and short-axis tumor measurements. FLAIR sequences were available for most, but not all patients. Few patients have FLAIR available at one time point and not at another. For scans with both T2-weighted and FLAIR sequences, readers measured either axial T2 or FLAIR images based on their preference.
These measurements were analyzed utilizing RAPNO imaging criteria to categorize the response. PD in the pons was defined as a ≥ 25% increase in the 2D product of the perpendicular diameters compared with the smallest measurement at any time-point using T2-weighted or FLAIR sequences. PR in the pons was defined as a ≥ 25% decrease in the 2D product of the largest perpendicular diameters compared with baseline measurement using T2-weighted or FLAIR sequences. CR was defined as no evidence of enhancing or nonenhancing, measurable or nonmeasurable disease, and no new lesions. SD was defined as a disease that did not meet the criteria for CR, PR, or PD.
Serial MRIs were evaluated for response. The response category for each MRI was compared between 2 readers. Their response was classified as concordant if they agreed on their response category, minor discordant if one reported SD and the other reported PR or PD, and major discordant if one reported PR and the other reported PD.
Results
Summary demographics on this cohort have been published previously.25 Briefly, the median patient age at the time of diagnosis was 6.8 years (range 2-27 years), 22 (48%) patients were female, and the median overall survival of the cohort was 15 months (Range: 5-34 months). All 46 patients received upfront radiation therapy, and 41 (89%) received systemic therapy (concurrent with radiation in 24 (52%) and adjuvant in 40 (87%)). A total of 312 MRIs were provided by the Registry, 35 MRI studies could not be projected on one or both readers’ Ambrahealth account, resulting in 277 MRIs from 46 patients being analyzed (Figure 1, Supplementary Data 1 and 2). There was a median of 6 MRIs per patient (range 0–13).
Figure 1.
Flow of the study participants (CONSORT) diagram. Among 312 paired MRIs that met the inclusion criteria, 10, 7, and 18 MRIs were not accessed by both readers, reader 1 and reader 2 respectively. A total of paired 277 MRIs were accessible by both readers and available for size measurements. Among these, 39 pairs of MRIs were not reviewed for a response as they were baseline (response category not applicable for baseline MRI), 21 pairs of MRIs were not reviewed for a response as their baseline MRI were not accessible (response category can’t be determined for a patient if baseline measurement is not available), and 93 MRIs were not reviewed for a response as these were post-PD determination in respective patients (once a patient is determined to have PD, the response type does not change). A total of 124 paired MRIs were reviewed for response.
Concordance:
124 paired MRIs were evaluable for concordance. The response category was concordant between 2 readers in 84 (68%) MRIs and discordant in the remaining 40 (32%) MRIs, including 31 (25%) comparisons that were minor discordant and 9 (7%) that were major discordant (Figure 2, and Table 1).
Figure 2.
Comparison of the response category among the 2 pediatric neuro-radiologists. Concordance between 2 pediatric neuro-radiologists and further breakdown of minor discordant cases into those 10% or less from the boundary of SD/PD and SD/PR vs those more than 10% from the boundary of SD/PD and SD/PR. PD: Progressive disease, SD: Stable disease, PR: Partial response.
Table 1.
Comparison of Response Type Between 2 Readers.
| CR by reader 1 | PR by reader 1 | SD by reader 1 | PD by reader 1 | |
|---|---|---|---|---|
| CR by reader 2 | 0 | 0 | 0 | 0 |
| PR by reader 2 | 0 | 27 | 12 | 2 |
| SD by reader 2 | 0 | 1 | 28 | 6 |
| PD by reader 2 | 0 | 7 | 12 | 29 |
Sustainability of PD
We tested the sustainability of PD with a follow-up MRI in 4-8 weeks. A 4–8-week follow-up MRI after determination of PD was available for 30 of those. For remaining patients, a follow-up MRI in 4-8 weeks after determination of PD was not available. PD was sustained in the follow-up MRI in 22 (73%) of cases and was not sustained in 8 (27%) of the cases. Out of 8 cases where PD was not sustained in a follow-up MRI, 7 (87.5%) were in the 3 months following radiation therapy, suggesting the influence of pseudo-progression. Only one case in which PD was not sustained was outside of the 3 months immediately following radiation therapy and therapy administered before PD was not known.
Discordant Cases
Minor discordant cases were the ones where one reader reported SD and the other reported PD or PR. Additional analysis of these cases showed that the measurements of the reader reporting PD or PR were 10% or less outside the boundary of the response category (SD/PD or SD/PR) in 20 (65%) of these cases (Figure 2).
We also investigated the cases with major discordance. The PR is reported based on a comparison to the baseline tumor measurement and the PD is reported based on a comparison to the smallest measurement to date. Because of different benchmarks for PD and PR, sometimes the responses of 2 readers had major discordance despite only small differences in measurements as shown in Figure 3.
Figure 3.
A case of major discordance between 2 readers: A: Baseline measurements by 2 readers, B: Smallest measurement to date and response type PR by 2 readers, 52% decrease in size compared to the baseline measurement by reader 1 and 64% decrease in size compared to the baseline by reader 2, C: Discordance response type by 2 readers, 18% increase in size compared to the smallest measurement and 44% decrease in size compared to the baseline by reader 1 and response type PR, 33% increase in size compared to the smallest measurement and 52% decrease in size compared to the baseline by reader 2 and response type PD. PD: Progressive disease, PR: Partial response.
Variations in Measurements
The 2 readers did not agree on the size of the tumor all the time. The median difference between the 2 readers’ long axis measurements was 2 mm, the mean difference in long axis measurement was 3.15 mm (range 0–27 mm), the median difference in short axis measurements was 2 mm, and the mean difference in short axis measurement was 3.25 mm (range 0–29 mm). Thirteen MRIs with the largest differences in measurements between 2 readers were remeasured by the same readers. The measurements changed by a small margin but continued to have a high degree of difference between the readers. There was no specific pattern to the difference in measurements between 2 readers. For some MRIs, one reader had a larger measurement of the 2, and for other MRIs, the other reader had a larger measurement of the 2 readers.
Discussion
The current study tests the clinical applicability of RAPNO DIPG working group imaging criteria for the primary pontine lesion. By applying the imaging component of the RAPNO DIPG working group criteria to a set of IDIPGR patients, we demonstrated concordance or minor discordance in 93% of the cases between 2 readers. (Figure 2).
DIPG is a very infiltrative tumor with indistinct borders, making tumor measurements challenging. The readers report seeing a transitional zone and edema around the tumor, and the predicament of whether to include them in the tumor measurement. The majority (65%) of the minor discordant cases were 10% or less from the boundary between SD/PR or SD/PD and 2 readers may have interpreted the transitional zone differently, resulting in a discrepancy in the size of the tumor and response category. A small difference in the measurement between 2 readers may result in a discordant response categorization, especially in a small tumor (Figure 3). While for this study, definitions of discordance were based on the contiguity of response categories between the 2 readers, the authors recognize that what is defined as “minor” discordance may have major therapeutic implications. For example, if the response category is PD from one reader and SD from the other reader, the patient with SD continues their treatment while the patient with PD may have to be removed from the study therapy. The median difference in the short and long-axis measurements between the 2 readers was 2 mm, and this small difference in measurement accounted for all the discordant response categories (Figure 4). Previous studies have shown large inter-observer variation in 1-, 2-, and 3-dimensional measurements of DIPG.26
Figure 4.
Graphic depiction of select cases with large differences in measurements between 2 readers. Measurements shown are long axis (mm), short axis (mm), and 2D products (mm2). A: 10.5 mm difference in the long axis and 14.4 mm difference in the short axis resulted in a 1377 mm2 difference in the 2D area, B: 8 mm difference in the long axis and 29 mm difference in the short axis resulted in a 2133 mm2 difference in the 2D area, C: 14 mm difference in the long axis and 2 mm difference in the short axis resulted in a 1000 mm2 difference in the 2D area.
We noted that even a small difference in the measurements between 2 readers may result major in discordance in response categories due to the different benchmarks used to define PR and PD (figure 3). Other response evaluation criteria like RECIST also use different benchmarks for PR and PD. Future updates in response evaluation criteria may address this issue.
The area is a continuous variable and creating ordinal variables like PD, SD, PR, and CR from a continuous variable has its challenges. For example, there must be a 25% decrease in the tumor size from the baseline measurement to be categorized as PR, and a few millimeters difference in the measurement may change the response category from partial response to stable disease. In one of the MRIs, 2 readers measured a 25% and 16% decrease in size, but the cutoff mark of 25% resulted in a response category of PR by one and SD by the other.
The discrepancy in the measurements by readers emphasizes the need for the MRIs of a patient with DIPG to be evaluated by the same reader side by side for consistency, and central radiology review in clinical trials. This will enhance the accuracy of measurements over time. These interpersonal differences can be further minimized using the computer-generated margins of the tumor using machine learning, but this needs to be studied in clinical trials for consistency and accuracy.27,28
The RAPNO DIPG criteria recommend testing for the sustainability of PD with a follow-up MRI in 4-8 weeks if the treating physician suspects pseudo-progression. This sustainability of PD is of utmost importance in the setting of clinical trials to avoid premature discontinuation of protocol treatment because of the pseudo-progression. In our analysis, most of the cases of PDs not sustained in a 4–8-week follow-up MRI (87.5%) were in the 3 months immediately following radiation therapy, underscoring the need to confirm the sustainability of PD in the immediate postradiation period. Only one patient who did not sustain PD in a follow-up MRI was outside the 3-month postradiation period, and it is not clear if this patient received any other treatment associated with pseudo-progression like immunotherapy before the PD determination.
This is a retrospective study with a limited sample size and subject to selection bias. Only 2 neuro-radiologists reviewed the MRIs for this study. Additional readers would likely support the measurements of one of the 2 neuro-radiologists when readings were discrepant. However, the primary objective of this analysis was to study the operative performance of the RAPNO DIPG criteria –two readers readily identified the utility and pitfalls of the criteria such that additional readers were not felt to be necessary for this retrospective analysis. The RAPNO criteria include clinical characteristics of the patient in addition to imaging characteristics like a neurological exam and use of concurrent medication like steroid or anti-angiogenic drugs which was not reviewed for this study. While this is a piece of very relevant clinical information, it does not impact this study as it was focused on the operative performance of the imaging component of the RAPNO DIPG criteria. The RAPNO DIPG criteria include recommendations for pontine and extra-pontine disease response evaluation. Because of the limitation of the dataset, the current analysis was limited to the pontine lesions, and metastatic lesions were excluded (MRI spine not available) from the aim of this study. The RAPNO DIPG criteria allows the images to be measured in the T2-weighted or FLAIR sequences. While every effort was made to be consistent with the sequences, it was not always possible. It may affect the accuracy of the measurements. RAPNO DIPG criteria recommend confirmation of response with a repeat MRI in 4-8 weeks. Because of the limitation of the dataset (follow-up MRI in 4-8 weeks not available for some cases), response confirmation or sustainability could not be tested in all cases.
In conclusion, the RAPNO DIPG imaging criteria can be applied with concordance/minor discordance between readers in almost all cases. When discordant, measurements were largely at the boundaries of response type. Major discordance was attributed to different benchmarks for PD and PR categories. The operative performance of imaging may be further augmented if reviewed in the context of relevant clinical information like neurological examination findings, concurrent medication use, and sustainability of the response.
Supplementary material
Supplementary material is available online at Neuro-Oncology Practice (https://academic.oup.com/nop/).
Acknowledgement
We are indebted to the children and families who have suffered from DIPG for their invaluable contribution to this research. For financial support of the IDIPGR, we thank The Cure Starts Now Foundation, The Cure Starts Now Australia, Brooke Healey Foundation, Wayland Villars Foundation, Aidan’s Avengers, Aubreigh’s Army, Austin Strong, Cure Brain Cancer, Jeffrey Thomas Hayden Foundation, Laurie’s Love Foundation, Love, Chloe Foundation, Musella Foundation, Pray Hope Believe, Reflections Of Grace, Storm the Heavens Fund, Whitley’s Wishes, Gabriella’s Smile Foundation, The Gold Hope Project, The Isabella and Marcus Foundation, Lauren’s Fight for Cure, Robert Connor Dawes Foundation, Ryan’s Hope, Benny’s World, Lily Larue Foundation, Marlee’s Mission, RUN DIPG, American Childhood Cancer Organization, The DIPG/DMG Collaborative, and Snapgrant.com. This work was previously presented as an abstract at ISPNO2024 Philadelphia, USA.
Contributor Information
Hamza S Gorsi, Department of Pediatrics, Central Michigan University, Mount Pleasant, MI; Division of Hematology/Oncology, Children’s Hospital of Michigan, Detroit, MI, USA.
Melike Guryildirim, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Michael Kuwabara, Phoenix Children Hospital, Phoenix, AZ, USA.
Jovan Dhatt, Midwestern University, Glendale, AZ, USA.
Lindsey M Hoffman, Phoenix Children Hospital, Phoenix, AZ, USA.
Kenneth J Cohen, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD, USA.
Funding
There was no direct funding received for this study.
Conflict of interest statement
HSG, MG, MK, JD, KJC have no relevant conflicts of interest to report. LMH participates in Pediatric Brain Tumor Consortium, Children’s Oncology Group and Georgia Cancer Center’s data safety monitoring board or advisory board, and is the chair of international DIPG/DMG Registry.
Authorship statement
Conception and design: HSG, LMH, KJC. Data acquisition: MG, MK, JD. Data analysis: HSG. Draft writing and review: All. Final approval: All.
Data Availability
Measurements by 2 readers are provided as supplemental documents.
References
- 1. Louis DN, Perry A, Wesseling P, et al. The 2021 WHO classification of tumors of the central nervous system: a summary. Neuro Oncol. 2021;23(8):1231–1251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Wu G, Diaz AK, Paugh BS, et al. The genomic landscape of diffuse intrinsic pontine glioma and pediatric non-brainstem high-grade glioma. Nat Genet. 2014;46(5):444–450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Khuong-Quang DA, Buczkowicz P, Rakopoulos P, et al. K27M mutation in histone H3.3 defines clinically and biologically distinct subgroups of pediatric diffuse intrinsic pontine gliomas. Acta Neuropathol. 2012;124(3):439–447. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Buczkowicz P, Hoeman C, Rakopoulos P, et al. Genomic analysis of diffuse intrinsic pontine gliomas identifies three molecular subgroups and recurrent activating ACVR1 mutations. Nat Genet. 2014;46(5):451–456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Buczkowicz P, Bartels U, Bouffet E, Becher O, Hawkins C.. Histopathological spectrum of paediatric diffuse intrinsic pontine glioma: diagnostic and therapeutic implications. Acta Neuropathol. 2014;128(4):573–581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Hoffman LM, Veldhuijzen van Zanten SEM, Colditz N, et al. Clinical, radiologic, pathologic, and molecular characteristics of long-term survivors of Diffuse Intrinsic Pontine Glioma (DIPG): a collaborative report from the international and European society for pediatric oncology DIPG registries. J Clin Oncol. 2018;36(19):1963–1972. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Cohen KJ, Heideman RL, Zhou T, et al. Temozolomide in the treatment of children with newly diagnosed diffuse intrinsic pontine gliomas: a report from the children’s oncology group. Neuro Oncol. 2011;13(4):410–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Chassot A, Canale S, Varlet P, et al. Radiotherapy with concurrent and adjuvant temozolomide in children with newly diagnosed diffuse intrinsic pontine glioma. J Neurooncol. 2012;106(2):399–407. [DOI] [PubMed] [Google Scholar]
- 9. Jalali R, Raut N, Arora B, et al. Prospective evaluation of radiotherapy with concurrent and adjuvant temozolomide in children with newly diagnosed diffuse intrinsic pontine glioma. Int J Radiat Oncol Biol Phys. 2010;77(1):113–118. [DOI] [PubMed] [Google Scholar]
- 10. Hargrave D, Bartels U, Bouffet E.. Diffuse brainstem glioma in children: critical review of clinical trials. Lancet Oncol. 2006;7(3):241–248. [DOI] [PubMed] [Google Scholar]
- 11. Osorio DS, Patel N, Ji L, et al. Pre-irradiation intensive induction and marrow-ablative consolidation chemotherapy in young children with newly diagnosed high-grade brainstem gliomas: report of the “head-start” I and II clinical trials. J Neurooncol. 2018;140(3):717–725. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Manley PE, Trippett T, Smith AA, et al. A phase 1/2 dose-finding, safety, and activity study of cabazitaxel in pediatric patients with refractory solid tumors including tumors of the central nervous system. Pediatr Blood Cancer. 2018;65(9):e27217. [DOI] [PubMed] [Google Scholar]
- 13. Bartels U, Wolff J, Gore L, et al. Phase 2 study of safety and efficacy of nimotuzumab in pediatric patients with progressive diffuse intrinsic pontine glioma. Neuro Oncol. 2014;16(11):1554–1559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Warren K, Bent R, Wolters PL, et al. A phase 2 study of pegylated interferon α-2b (PEG-Intron(®)) in children with diffuse intrinsic pontine glioma. Cancer. 2012;118(14):3607–3613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Baxter PA, Su JM, Onar-Thomas A, et al. A phase I/II study of veliparib (ABT-888) with radiation and temozolomide in newly diagnosed diffuse pontine glioma: a pediatric brain tumor consortium study. Neuro Oncol. 2020;22(6):875–885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Su JM, Kilburn LB, Mansur DB, et al. Phase I/II trial of vorinostat and radiation and maintenance vorinostat in children with diffuse intrinsic pontine glioma: a children’s oncology group report. Neuro Oncol. 2022;24(4):655–664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Warren KE, Poussaint TY, Vezina G, et al. Challenges with defining response to antitumor agents in pediatric neuro-oncology: a report from the response assessment in pediatric neuro-oncology (RAPNO) working group. Pediatr Blood Cancer. 2013;60(9):1397–1401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Cooney TM, Cohen KJ, Guimaraes CV, et al. Response assessment in diffuse intrinsic pontine glioma: recommendations from the Response Assessment in Pediatric Neuro-Oncology (RAPNO) working group. Lancet Oncol. 2020;21(6):e330–e336. [DOI] [PubMed] [Google Scholar]
- 19. Hoffman LM, Jaimes C, Mankad K, et al. Response assessment in pediatric craniopharyngioma: recommendations from the Response Assessment in Pediatric Neuro-Oncology (RAPNO) WORKING GROUP. Neuro Oncol. 2023;25(2):224–233. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Warren KE, Vezina G, Poussaint TY, et al. Response assessment in medulloblastoma and leptomeningeal seeding tumors: recommendations from the response assessment in pediatric neuro-oncology committee. Neuro Oncol. 2018;20(1):13–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Lindsay HB, Massimino M, Avula S, et al. Response assessment in paediatric intracranial ependymoma: recommendations from the Response Assessment in Pediatric Neuro-Oncology (RAPNO) working group. Lancet Oncol. 2022;23(8):e393–e401. [DOI] [PubMed] [Google Scholar]
- 22. Fangusaro J, Witt O, Hernáiz Driever P, et al. Response assessment in paediatric low-grade glioma: recommendations from the Response Assessment in Pediatric Neuro-Oncology (RAPNO) working group. Lancet Oncol. 2020;21(6):e305–e316. [DOI] [PubMed] [Google Scholar]
- 23. Erker C, Tamrazi B, Poussaint TY, et al. Response assessment in paediatric high-grade glioma: recommendations from the Response Assessment in Pediatric Neuro-Oncology (RAPNO) working group. Lancet Oncol. 2020;21(6):e317–e329. [DOI] [PubMed] [Google Scholar]
- 24. Baugh J, Bartels U, Leach J, et al. The international diffuse intrinsic pontine glioma registry: an infrastructure to accelerate collaborative research for an orphan disease. J Neurooncol. 2017;132(2):323–331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Lazow MA, Nievelstein MT, Lane A, et al. Volumetric endpoints in diffuse intrinsic pontine glioma: comparison to cross-sectional measures and outcome correlations in the International DIPG/DMG Registry. Neuro Oncol. 2022;24(9):1598–1608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Hayward RM, Patronas N, Baker EH, et al. Inter-observer variability in the measurement of diffuse intrinsic pontine gliomas. J Neurooncol. 2008;90(1):57–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Familiar AM, Fathi Kazerooni A, Vossough A, et al. Towards consistency in pediatric brain tumor measurements: challenges, solutions, and the role of artificial intelligence-based segmentation. Neuro Oncol. 2024;26(9):1557–1571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Fathi Kazerooni A, Arif S, Madhogarhia R, et al. Automated tumor segmentation and brain tissue extraction from multiparametric MRI of pediatric brain tumors: a multi-institutional study. Neurooncol. Adv.. 2023;5(1):vdad027. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Measurements by 2 readers are provided as supplemental documents.