ESTRO 2023

Session Item

Sunday

May 14

09:00 - 10:00

Stolz 1

Paediatrics - AYA

Chair: , Germany;

Co-Chair: Daniella Elisabet Østergaard, Denmark

Session Code: 2160

Session Type: Mini-Oral

Track: Clinical

(A)symptomatic MRI changes after proton therapy; the Dutch national cohort of pediatric CNS tumors

Jikke Rutgers , The Netherlands

Presentation Number: MO-0381

Abstract

Abstract Title:

(A)symptomatic MRI changes after proton therapy; the Dutch national cohort of pediatric CNS tumors

Authors:

Jikke Rutgers¹, Agata Bannink-Gawryszuk^2,1, Peter van Rossum³, Maarten Lequin^1,4, John Maduro^1,5, Hiske van der Weide^1,5, Mart Heesters^1,5, Bianca Hoeben^1,3, Astrid van der Heide¹, Sabine Plasschaert¹, Corrie Gidding¹, Evelien de Vos-Kerkhof¹, Lisethe Meijer¹, Jasper van der Lugt¹, Antoinette Schouten- van Meeteren¹, Witold Matysiak⁵, Enrica Seravalli³, Hans Langendijk⁵, Eelco Hoving¹, Geert Janssens^1,3

¹Princess Máxima Center for pediatric oncology, Pediatric oncology, Utrecht, The Netherlands; ²University Medical Center Groningen, Radiation oncology, Groningen , The Netherlands; ³University Medical Center Utrecht, Radiation oncology, Utrecht, The Netherlands; ⁴University Medical Center Utrecht, Radiology, Utrecht, The Netherlands; ⁵University Medical Center Groningen, Radiation oncology, Groningen, The Netherlands

Show Affiliations

Purpose or Objective

Imaging changes after proton therapy (PT) for CNS tumors, their clinical relevance and risk factors in children gained focus in literature. The aim of this study was to assess (a)symptomatic imaging changes after pencil beam scanning PT in the Dutch cohort of pediatric CNS tumors with a centralized follow-up.

Material and Methods

From the 100 pediatric patients with a newly diagnosed CNS tumor receiving PT at the University Medical Center Groningen between 07-2018 and 07-2021, 78 patients were eligible for retrospective analysis. Reasons for excluding 22 patients were: no informed consent, re-irradiation or a combination of PT with >2 fractions of photon therapy.

In line with international protocols, all clinical and radiological follow-up was performed on a regular basis at the Princess Màxima Center for Pediatric Oncology using uniform scan protocols. PT-related imaging changes were defined as new abnormal T2 signal with contrast enhancement on T1 (T1gd+) outside any potential macroscopic residual tumor, within the irradiated volume, and not suspicious for progressive disease. The relationship between imaging changes and clinical symptoms was discussed multidisciplinary with the treating neuro-oncologist/neurologist, two radiation oncologists and a (neuro)radiologist.

Kaplan-Meier and Cox proportional hazard models were used to calculate the cumulative incidence rates of (a)symptomatic imaging changes, and to evaluate potentially associated factors such as: gender, age, tumor location, chemotherapy, hydrocephalus, number of neurosurgical procedures, and dose to the craniospinal axis.

Results

A total of 78 patients and 556 MRI-exams were analyzed. Median PT dose was 54.0 GyE (range: 24.0-60.0) at a median age of 7.3 years (range: 1.1-18.9) and median follow-up from end of PT of 23.6 months (IQR: 14.8-36.8).

At 3, 6, 9 and 12 months from end of PT, the respective cumulative incidences of T1gd+ changes were 6%, 26%, 33%, and 33% while the prevalence rate at 3, 6, 9, 12, 15 and 18 months (from end of PT) was 7%, 28%, 25%, 24%, 17% and 10%. The cumulative incidence of symptomatic imaging changes at 3, 6, 9 and 12 months from end of PT was 1%, 10%, 13%, and 13%, respectively.

Patient age ≥8 versus <8 years significantly decreased the risk of T1gd+ changes (HR 0.15, 95% CI: 0.05-0.45, p=<0.01) with a cumulative incidence at 12 months of 11% vs. 58%, respectively. Older patients (≥8 years) were less susceptible to symptomatic imaging changes (HR:0.12, 95% CI:0.01-0.93, p=0.04) having a cumulative incidence at 12 months of 3% vs. 22%.

Conclusion

The incidence of PT-related MRI changes, based on centralized and structured clinical and radiological follow-up data of children with a CNS tumor treated in the Netherlands, has demonstrated T1gd+ changes in one third of patients. An increased risk of symptomatic imaging changes appears to be present in younger children.