A Prospective Study of Machine Learning-Assisted Radiation Therapy Planning for Patients Receiving 54 Gy to the Brain.

PURPOSE: The capacity for machine learning (ML) to facilitate radiation therapy (RT) planning for primary brain tumors has not been described. We evaluated ML-assisted RT planning with regard to clinical acceptability, dosimetric outcomes, and planning efficiency for adults and children with primary brain tumors. METHODS AND MATERIALS: In this prospective study, children and adults receiving 54 Gy fractionated RT for a primary brain tumor were enrolled. For each patient, one ML-assisted RT plan was created and compared with 1 or 2 plans created using standard ("manual") planning procedures. Plans were evaluated by the treating oncologist, who was blinded to the method of plan creation. The primary endpoint was the proportion of ML plans that were clinically acceptable for treatment. Secondary endpoints included the frequency with which ML plans were selected as preferable for treatment, and dosimetric differences between ML and manual plans. RESULTS: A total of 116 manual plans and 61 ML plans were evaluated across 61 patients. Ninety-four percent of ML plans and 93% of manual plans were judged to be clinically acceptable (P = 1.0). Overall, the quality of ML plans was similar to manual plans. ML plans comprised 34.5% of all plans evaluated and were selected for treatment in 36.1% of cases (P = .82). Similar tumor target coverage was achieved between both planning methods. Normal brain (brain minus planning target volume) received an average of 1 Gy less mean dose with ML plans (compared with manual plans, P < .001). ML plans required an average of 45.8 minutes less time to create, compared with manual plans (P < .001). CONCLUSIONS: ML-assisted automated planning creates high-quality plans for patients with brain tumors, including children. Plans created with ML assistance delivered slightly less dose to normal brain tissues and can be designed in less time.

Paediatric radiation therapy without anaesthesia - Are the children moving?

PURPOSE: Children who require radiation therapy (RT) should ideally be treated awake, without anaesthesia, if possible. Audiovisual distraction is a known method to facilitate awake treatment, but its effectiveness at keeping children from moving during treatment is not known. The aim of this study was to evaluate intrafraction movement of children receiving RT while awake. METHODS: In this prospective study, we measured the intrafraction movement of children undergoing treatment with fractionated RT, using pre- and post-RT cone beam CT (CBCT) with image matching on bony anatomy. Study CBCTs were acquired at first fraction, weekly during RT, and at last fraction. The primary endpoint was the magnitude of vector change between the pre- and post-RT scans. Our hypothesis was that 90 % of CBCT acquisitions would have minimal movement, defined as <3 mm for head-and-neck (HN) treatments and <5 mm for non-HN treatments. RESULTS: A total of 65 children were enrolled and had evaluable data across 302 treatments with CBCT acquisitions. Median age was 11 years (range, 2-18; 1st and 3rd quartiles 7 and 14 years, respectively). Minimal movement was observed in 99.4 % of HN treatments and 97.2 % of non-HN treatments. The study hypothesis of >90 % of evaluations having minimal movement was met. Children who were age >11 years moved less at initial evaluation but tended to move more as a course of radiation progressed, as compared to children who were younger. CONCLUSION: Children receiving RT with audiovisual distraction while awake had small magnitudes of observed intrafraction movement, with minimal movement in >97 % of observed RT fractions. This study validates methods of anaesthesia avoidance using audiovisual distraction for selected children.

Late Cardiac Toxic Effects Associated With Treatment Protocols for Hodgkin Lymphoma in Children.

Importance: Contemporary North American trials for children with Hodgkin lymphoma (HL) have decreased radiation therapy (RT) use and increased pharmacologic cardioprotection but also increased the cumulative doxorubicin dose, making overall treatment consequences for late cardiac toxic effects uncertain. Objective: To estimate the risk of cardiac toxic effects associated with treatments used in modern pediatric HL clinical trials. Design, Setting, and Participants: For this cohort study, Fine and Gray models were fitted using survivors in the Childhood Cancer Survivor Study who were diagnosed with HL between January 1, 1970, and December 31, 1999, and were followed for a median of 23.5 (range, 5.0-46.3) years. These models were applied to the exposures in the study population to estimate the 30-year cumulative incidence of cardiac disease. The study population comprised patients with intermediate-risk or high-risk HL treated in 4 consecutive Children's Oncology Group clinical trials from September 2002 to October 2022: AHOD0031, AHOD0831, AHOD1331, and S1826. Data analysis was performed from April 2020 to February 2023. Exposures: All patients received chemotherapy including doxorubicin, and some patients received mediastinal RT, dexrazoxane, or mediastinal RT and dexrazoxane. Main Outcomes and Measures: Estimated 30-year cumulative incidence of grade 3 to 5 cardiac disease. Results: The study cohort comprised 2563 patients, with a median age at diagnosis of 15 (range, 1-22) years. More than half of the patients were male (1357 [52.9%]). All 2563 patients received doxorubicin, 1362 patients (53.1%) received mediastinal RT, and 307 patients (12.0%) received dexrazoxane. Radiation therapy use and the median mean heart dose among patients receiving RT decreased, whereas the planned cumulative dose of doxorubicin and use of dexrazoxane cardioprotection increased. For patients treated at age 15 years, the estimated 30-year cumulative incidence of severe or fatal cardiac disease was 9.6% (95% CI, 4.2%-16.4%) in the AHOD0031 standard treatment group (enrolled 2002-2009), 8.6% (95% CI, 3.8%-14.9%) in the AHOD0831 trial (enrolled 2009-2012), 8.2% (95% CI, 3.6%-14.3%) in the AHOD1331 trial (enrolled 2015-2019), and 6.2% (95% CI, 2.7%-10.9%) in the S1826 trial (enrolled 2019-2022), whereas the expected rate in an untreated population was 5.0% (95% CI, 2.1%-9.3%). Despite the estimated reduction in late cardiac morbidity, the frequency of recommended echocardiographic screening among survivors will increase based on current guidelines. Conclusions and Relevance: In this cohort study of sequential HL trials, reductions in the proportion of children receiving mediastinal RT and increases in dexrazoxane use were estimated to offset the increased doxorubicin dose and produce a net reduction in late cardiac disease. Further studies on dexrazoxane are warranted to confirm whether its role in reducing cardiac toxic effects is maintained long term. These findings suggest that survivorship follow-up guidelines should be refined to align with the risks associated with treatment.

Proton Therapy Mediates Dose Reductions to Brain Structures Associated With Cognition in Children With Medulloblastoma.

PURPOSE: Emerging evidence suggests proton radiation therapy may offer cognitive sparing advantages over photon radiation therapy, yet dosimetry has not been compared previously. The purpose of this study was to examine dosimetric correlates of cognitive outcomes in children with medulloblastoma treated with proton versus photon radiation therapy. METHODS AND MATERIALS: In this retrospective, bi-institutional study, dosimetric and cognitive data from 75 patients (39 photon and 36 proton) were analyzed. Doses to brain structures were compared between treatment modalities. Linear mixed-effects models were used to create models of global IQ and cognitive domain scores. RESULTS: The mean dose and dose to 40% of the brain (D40) were 2.7 and 4.1 Gy less among proton-treated patients compared with photon-treated patients (P = .03 and .007, respectively). Mean doses to the left and right hippocampi were 11.2 Gy lower among proton-treated patients (P < .001 for both). Mean doses to the left and right temporal lobes were 6.9 and 7.1 Gy lower with proton treatment, respectively (P < .001 for both). Models of cognition found statistically significant associations between higher mean brain dose and reduced verbal comprehension, increased right temporal lobe D40 with reduced perceptual reasoning, and greater left temporal mean dose with reduced working memory. Higher brain D40 was associated with reduced processing speed and global IQ scores. CONCLUSIONS: Proton therapy reduces doses to normal brain structures compared with photon treatment. This leads to reduced cognitive decline after radiation therapy across multiple intellectual endpoints. Proton therapy should be offered to children receiving radiation for medulloblastoma.

Estimating Potential Benefits to Neurocognition with Proton Therapy in Adults with Brain Tumors.

Purpose: Photon radiation therapy (RT) is important in the treatment of many brain tumors but can negatively affect neurocognition. Proton therapy (PT) can reduce doses to normal brain structures. We compared photon and proton plans to estimate the potential benefit in cognition if the patient were treated with PT. Materials and Methods: We analyzed 23 adult patients with proton and photon plans for the treatment of a primary brain tumor. Cognitive outcomes were predicted using converted equivalent dose (EQD2) with an α/ß ratio of 3 to left temporal lobe and normal brain tissue. Risks of cognitive decline on 2 specific tests, the Controlled Oral Word Association Test (COWAT [letter S], a test of verbal fluency) and the Wechler Adult Intelligence Scale (WAIS-IV Coding Test, a test of processing speed) were derived from a previously published model. Results: Dose reductions to left temporal lobe and normal brain tissue translated into lower estimated probabilities of impairment in specific neurocognitive test scores after PT. With a mean dose reduction from 1490 to 1092 cGy in EQD2 to the left temporal lobe (P < .001), there was reduction in probability of impairment in the COWAT (Letter S) test from 6.8% to 5.4%. Similar results were seen with the normal brain (750 to 451 cGy in EQD2, P < .001), with reduction in probability of impairment in the WAIS-IV Coding test from 5% to 4.1%. Other structures experiencing dose reduction with PT included each cochlea, posterior fossa, each temporal lobe, and each hippocampus. Conclusion: We confirmed an association between PT and lower doses to brain substructures, which is expected to result in a modest decrease in probability of impairment in neurocognitive test scoring. These findings should be confirmed in prospective cohorts of patients treated with PT.

Photon versus proton whole ventricular radiotherapy for non-germinomatous germ cell tumors: A report from the Children's Oncology Group.

PURPOSE: To determine if proton therapy reduces doses to cranial organs at risk (OARs) as compared to photon therapy in children with non-germinomatous germ cell tumors (NGGCT) receiving whole ventricular radiotherapy (WVRT). METHODS AND MATERIALS: Dosimetric data for patients with NGGCT prospectively enrolled in stratum 1 of the Children's Oncology Group study ACNS1123 who received 30.6 Gy WVRT were compared. Target segmentation was standardized using a contouring atlas. Doses to cranial OARs were compared between proton and photon treatments. Clinically relevant dose-volume parameters that were analyzed included mean dose and dose to 40% of the OAR volume (D40). RESULTS: Mean and D40 doses to the supratentorial brain, cerebellum, and bilateral temporal, parietal, and frontal lobes were statistically significantly lower amongst proton-treated patients, as compared to photon-treated patients. In a subgroup analysis of patients uniformly treated with a 3-mm planning target volume, patients who received proton therapy continued to have statistically significantly lower doses to brain OARs. CONCLUSIONS: Children treated with proton therapy for WVRT had lower doses to normal brain structures, when compared to those treated with photon therapy. Proton therapy should be considered for patients receiving WVRT for NGGCT.

A pilot study of machine-learning based automated planning for primary brain tumours.

PURPOSE: High-quality radiotherapy (RT) planning for children and young adults with primary brain tumours is essential to minimize the risk of late treatment effects. The feasibility of using automated machine-learning (ML) to aid RT planning in this population has not previously been studied. METHODS AND MATERIALS: We developed a ML model that identifies learned relationships between image features and expected dose in a training set of 95 patients with a primary brain tumour treated with focal radiotherapy to a dose of 54 Gy in 30 fractions. This ML method was then used to create predicted dose distributions for 15 previously-treated brain tumour patients across two institutions, as a testing set. Dosimetry to target volumes and organs-at-risk (OARs) were compared between the clinically-delivered (human-generated) plans versus the ML plans. RESULTS: The ML method was able to create deliverable plans in all 15 patients in the testing set. All ML plans were generated within 30 min of initiating planning. Planning target volume coverage with 95% of the prescription dose was attained in all plans. OAR doses were similar across most structures evaluated; mean doses to brain and left temporal lobe were lower in ML plans than manual plans (mean difference to left temporal, - 2.3 Gy, p = 0.006; mean differences to brain, - 1.3 Gy, p = 0.017), whereas mean doses to right cochlea and lenses were higher in ML plans (+ 1.6-2.2 Gy, p < 0.05 for each). CONCLUSIONS: Use of an automated ML method to aid RT planning for children and young adults with primary brain tumours is dosimetrically feasible and can be successfully used to create high-quality 54 Gy RT plans. Further evaluation after clinical implementation is planned.

Intellectual changes after radiation for children with brain tumors: which brain structures are most important?

BACKGROUND: The objective of this study was to evaluate the contribution of radiation dose to different intracranial structures on changes in intellectual function for children with brain tumors. METHODS: We evaluated children with brain tumors treated in 2005-2017 who had longitudinal neuropsychological assessments and available photon dosimetric data (if radiation therapy [RT] given). Full Scale Intelligence Quotient (FSIQ) and index scores were evaluated (perceptual reasoning index [PRI], processing speed index [PSI], verbal comprehension index [VCI], and working memory index [WMI]). Multivariable linear mixed effects models were used to model endpoints, with age at RT and dose to different brain regions as fixed effects and patient-specific random intercepts. P-values (P*) were adjusted for multiple comparisons. RESULTS: Sixty-nine patients were included, 56 of whom received RT. Median neuropsychological follow-up was 3.2 years. Right temporal lobe mean dose was strongly associated with decline in FSIQ (P* = 0.005); with each gray increase in mean dose, there was a decrease of 0.052 FSIQ points per year. Dose to 50% (D50) of the supratentorial brain was associated with decline in PSI (P* = 0.006) and WMI (P* = 0.001). Right and left hippocampus D50 were individually strongly associated with declines in VCI (P* = 0.009 for each). Presence of a ventriculoperitoneal shunt decreased FSIQ by 10 points. CONCLUSIONS: We reported associations between dosimetry to specific brain regions and intellectual outcomes, with suggested avoidance structures during RT planning. These models can help clinicians anticipate changes in neurocognition post-RT and guide selection of an optimal RT plan.