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

PURPOSE/OBJECTIVE(S)

Radiotherapy (RT) planning is presently a semi-manual, iterative, labor-intensive process which may result in unnecessary variation in plan quality. To improve treatment plan quality and decrease RT planning time, we conducted a prospective, blinded study to compare machine learning-assisted planning with conventional manual planning for patients receiving 54 Gy in 30 fractions for a primary brain tumor.

MATERIALS/METHODS

From January 31, 2022 to January 10, 2023, 40 patients receiving 54 Gy for primary CNS tumors were prospectively enrolled (median age 50 years, range 4-78 years). Patients underwent standard CT/MR simulation and target/OAR delineation by the treating radiation oncologist. Each patient had one ML plan and 1-2 manual RT plans created by different planners. The reviewing oncologist was blinded to planning method by removing optimization and IMRT/VMAT beam arrangement details from all plans, which were then rated based on clinical acceptability, target coverage, OAR sparing, conformity, and dose-fall off. One preferred plan was chosen and used for clinical treatment.

RESULTS

A total of 115 plans for 40 patients were evaluated: 40 ML plans (35% of all plans), and 75 manual plans (65% of all plans; 5 and 35 patients had 1 and 2 manual plans created, respectively). ML plans required a mean planning time of 65 min as compared to 107 min for manual plans, with a mean time savings of 41 min per patient (paired t-test p = 0.002). 97% of ML plans (95% confidence interval [CI] 85-100) and 96% of manual plans (95% CI 87-99) were designated clinically acceptable by the treating radiation oncologist. While ML-assisted plans represented 35% of plans evaluated, they were chosen as preferred for clinical treatment in 43% of cases (17/40, 95% CI 29-58, p = 0.32). Median doses to the brain (10.8 Gy vs. 11.3 Gy, Wilcoxon rank-sum p = 0.012) and brain minus PTV (9.2 Gy vs 10.0 Gy, Wilcoxon rank-sum p = 0.009) were lower with ML planning versus manual planning, respectively. Doses to other structures, including hippocampi, cochlea, pituitary and hypothalamus were not statistically different.

CONCLUSION

In this prospective study with blinded oncologist evaluation, ML-assisted RT planning for primary CNS tumors was faster than manual planning, and produced a very high rate of acceptable plans with similar or superior OAR sparing. Future work will be undertaken to iteratively refine the ML model using the preferred cases from this study.

The Brain Metastases Symptom Checklist as a novel tool for symptom measurement in patients with brain metastases undergoing whole-brain radiotherapy

PURPOSE

We evaluated the feasibility, reliability, and validity of the Brain Metastases Symptom Checklist (bmsc), a novel self-report measure of common symptoms experienced by patients with brain metastases.

METHODS

Patients with first-presentation symptomatic brain metastases (n = 137) referred for whole-brain radiotherapy (wbrt) completed the bmsc at time points before and after treatment. Their caregivers (n = 48) provided proxy ratings twice on the day of consultation to assess reliability, and at week 4 after wbrt to assess responsiveness to change. Correlations with 4 other validated assessment tools were evaluated.

RESULTS

The symptoms reported on the bmsc were largely mild to moderate, with tiredness (71%) and difficulties with balance (61%) reported most commonly at baseline. Test-retest reliability for individual symptoms had a median intraclass correlation of 0.59 (range: 0.23-0.85). Caregiver proxy and patient responses had a median intraclass correlation of 0.52. Correlation of absolute scores on the bmsc and other symptom assessment tools was low, but consistency in the direction of symptom change was observed. At week 4, change in symptoms was variable, with improvements in weight gain and sleep of 42% and 41% respectively, and worsening of tiredness and drowsiness of 62% and 59% respectively.

CONCLUSIONS

The bmsc captures a wide range of symptoms experienced by patients with brain metastases, and it is sensitive to change. It demonstrated adequate test-retest reliability and face validity in terms of its responsiveness to change. Future research is needed to determine whether modifications to the bmsc itself or correlation with more symptom-specific measures will enhance validity.

Treatment Planning and Delivery Evaluation of Volumetric Modulated Arc Therapy for Stereotactic Body Radiotherapy of Spinal Tumours: Impact of Arc Discretization in Planning System

The purpose of this study was to compare single arc volumetric modulated arc therapy (VMAT) to intensity modulated radiotherapy (IMRT) for spine SBRT in terms of target coverage, organ at risk (OAR) sparing and delivery performance. VMAT plans with 91 control points (VMAT-91CP) were generated for 15 spine metastases patients previously treated with a nine-field IMRT technique. VMAT and IMRT plans were compared based on target coverage, maximum spinal cord dose, maximum plan dose and volume of normal tissue receiving 20% to 80% of the prescribed dose. Treatment delivery time and monitor units (MU) were measured to determine delivery efficiency. To assess the impact of arc discretization in the treatment planning system (TPS), the VMAT-91CP plans were modified by almost doubling the number of CPs (VMAT-181CP). Planned-to-delivered dose agreement for both techniques was assessed using two types of 3D detector arrays. VMAT-91CP target coverage was equivalent to IMRT while maintaining or improving spinal cord sparing. This was achieved without increasing the volume of normal tissue receiving low or intermediate dose levels. Planned-to-delivered dose agreement equivalent to IMRT was achieved with VMAT, but required decreasing the CP angular spacing from 4° to 2° (VMAT-181CP plans). On average, VMAT-181CP plans reduced delivery time by 53% compared to IMRT. Single-arc VMAT for spine SBRT improved delivery efficiency while maintaining target coverage and OAR sparing compared to IMRT. VMAT plans generated with a CP gantry angular spacing of 2° is recommended to avoid a discretization effect in the TPS and ensure acceptable planned-to-delivered dose agreement.

Focal Stereotactic External Beam Radiotherapy as a Vision-sparing Method for the Treatment of Peripapillary and Perimacular Retinoblastoma: Preliminary Results

AIMS

Chemotherapy with aggressive focal ablative therapy is now the mainstay of retinoblastoma therapy. Our experience presents an evolution from conventional radiotherapy by treating posterior pole tumours with focal stereotactic fractionated radiotherapy (SRT).

MATERIALS AND METHODS

A retrospective chart review was conducted of five patients (six eyes) treated with SRT at the Hospital for Sick Children and Princess Margaret Hospital, Toronto, Canada, between 1999 and 2004. The prescribed dose was 40 Gy delivered in 20 fractions once daily using 6 MV photons.

RESULTS

Five patients (six eyes) were treated. The median age at the time of SRT was 18 months. The median follow-up was 46.5 months as of September 2004. Four patients were treated for a posterior pole focal tumour by focal SRT, and one patient was treated for vitreous seeding with whole-eye SRT. In patients treated with focal SRT, the median doses to the tumour, optic chiasm and brainstem were 41.92, 0.25 and 0.07 Gy, respectively, and to the ipsilateral optic nerve, globe and lens were 9.98, 19.11 and 3.74 Gy, respectively. The median doses to the ipsilateral and contralateral orbital bone were 6.73 Gy (range 5.99-8.29 Gy) and 2.31 Gy (range 0.88-7.08 Gy), respectively. A complete response (residual inactive scar tissue) was seen in four of the five focal tumours treated, with one tumour responding with a partial response (suspicious residual scar tissue). No acute or late side-effects occurred in patients treated with focal SRT. Only the patient treated with whole-eye SRT developed late effects of cataract and corneal ulceration. One patient suffered recurrence within the radiation field 5 months after focal SRT. Control of this recurrence was successful using chemotherapy and focal therapy. No eye has been enucleated.

CONCLUSION

Vision-sparing focal SRT for localised tumour masses in critical locations can control tumours with minimal side-effects and a minimal dose to the surrounding critical normal tissue.

Thyroid cancer management in North Trent. Case record review confirms the need for specialist care

Background

The Calman–Hine Report mandates the need for specialist care of patients with cancer within the UK. The next round of accreditation will include an assessment of care available to patients with thyroid cancer. The Joint Thyroid Cancer Clinic in Sheffield allows secondary and tertiary referral to a multidisciplinary specialist team for the two million people of North Trent.

Methods

A retrospective case record review of the last 159 patients referred to the clinic by specialist and non-specialist surgeons was performed to assess surgical and pathological aspects of care, including preoperative, peroperative and postoperative management, as well as operative morbidity.

Results

A total of 37 surgeons at ten hospitals were involved in the initial management of 128 women and 31 men (mean age 53 (range 10–99) years) with thyroid cancer (111 differentiated, nine medullary, 16 anaplastic, 15 lymphoma, eight others), including 23 general surgeons (69 patients), ten ear, nose and throat surgeons (ten patients) and three specialist endocrine surgeons (66 patients). Non-specialist surgeons carried out 78 reoperative procedures. Histopathological review after referral by a specialist pathologist resulted in a change in diagnosis in 10 per cent of 125 patients (13 cases). Operation-specific surgical morbidity was recorded as follows:

Cancer Registry data indicated that a further 101 patients with thyroid cancer were never referred to the Joint Thyroid Cancer Clinic during the same period.

Conclusion

In North Trent, despite the presence of a longstanding and well established clinic, many patients with thyroid cancer do not receive specialist/multidisciplinary care. Patients are often treated by surgeons with little experience of thyroid cancer, and the rates of inaccurate pathological diagnosis and operative morbidity are unacceptably high in patients treated by non-specialist teams.