Synthetic MRI-Assisted and Self-Supervised Adaptive Segmentation of Organs-at-Risk (OARs) in MRI-Based Radiation Therapy

PURPOSE/OBJECTIVE(S)

This study proposes a self-supervised solution for OAR segmentation, combining patch-based adaptation and unsupervised synthesis of T2-weighted MRI data to finetune the segmentation model. The aim is to improve adaptation to patient anatomy, overcome limited annotated MRI data, and enhance the generalizability of automatic segmentation models for gynecological cancers.

MATERIALS/METHODS

The study used a patch-based cycle consistent generative adversarial network (cycle-GAN) for unsupervised MRI synthesis from CT scans of 20 patients, and a residual U-Net model for OARs segmentation. The segmentation model was trained and validated on synthetic MRI (sMRI) of 103 and 25 patient scans respectively, then finetuned on 78 MRI scans from radiation therapy fractions of 15 additional patients through three-fold cross validation. Self-supervised adaptation was applied, incorporating affine and elastic deformations, intensity shifting, and scaling. The model was trained on 96 × 96 × 96 sub-volumes and validated on entire pelvic sections of the same images. A combination of Dice and weighted cross entropy (CE) losses, with weights assigned for bladder (1), small bowel (1), rectum (2), sigmoid (2), left femoral head (0), and right femoral head (0), was used for OAR segmentation. The performance was evaluated against the model trained only on a limited number of acquired MRI data, as well as sMRI pretrained models with encoder weight freezing and either equal weighting or soft-tissue adjusted weighting.

RESULTS

Our sMRI-assisted approach showed improved performance for challenging pelvic OARs compared to the method using only the acquired MRI data. The self-supervised fraction-adaptive segmentation results indicated better performance in target soft-tissues when using at least one treatment fraction for organ-specific adaptation.

CONCLUSION

Our framework leverages pre-existing CT planning data for gynecological cancers to enhance the segmentation performance of OARs during MR-guided adaptive treatments. This approach offers substantial benefits for the radiation therapy workflow, including reduced variability in per-fraction segmentation and clinical burden. Further studies that involve human expert evaluations will be conducted to assess the practicality of this approach in radiation therapy.

Understanding the Benefit of Magnetic Resonance-guided Adaptive Radiotherapy in Rectal Cancer Patients: a Single-centre Study

AIMS

Neoadjuvant chemoradiotherapy followed by surgery is the mainstay of treatment for patients with rectal cancer. Standard clinical target volume (CTV) to planning target volume (PTV) margins of 10 mm are used to accommodate inter- and intrafraction motion of target. Treating on magnetic resonance-integrated linear accelerators (MR-linacs) allows for online manual recontouring and adaptation (MRgART) enabling the reduction of PTV margins. The aim of this study was to investigate motion of the primary CTV (CTVA; gross tumour volume and macroscopic nodes with 10 mm expansion to cover microscopic disease) in order to develop a simultaneous integrated boost protocol for use on MR-linacs.

MATERIALS AND METHODS

Patients suitable for neoadjuvant chemoradiotherapy were recruited for treatment on MR-linac using a two-phase technique; only the five phase 1 fractions on MR-linac were used for analysis. Intrafraction motion of CTVA was measured between pre-treatment and post-treatment MRI scans. In MRgART, isotropically expanded pre-treatment PTV margins from 1 to 10 mm were rigidly propagated to post-treatment MRI to determine overlap with 95% of CTVA. The PTV margin was considered acceptable if overlap was >95% in 90% of fractions. To understand the benefit of MRgART, the same methodology was repeated using a reference computed tomography planning scan for pre-treatment imaging.

RESULTS

In total, nine patients were recruited between January 2018 and December 2020 with T3a-T4, N0-N2, M0 disease. Forty-five fractions were analysed in total. The median motion across all planes was 0 mm, demonstrating minimal intrafraction motion. A PTV margin of 3 and 5mm was found to be acceptable in 96 and 98% of fractions, respectively. When comparing to the computed tomography reference scan, the analysis found that PTV margins to 5 and 10 mm only acceptably covered 51 and 76% of fractions, respectively.

CONCLUSION

PTV margins can be reduced to 3-5 mm in MRgART for rectal cancer treatment on MR-linac within an simultaneous integrated boost protocol.

Feasibility of tumour-focused adaptive radiotherapy for bladder cancer on the MR-linac

Bladder tumour-focused magnetic resonance image-guided adaptive radiotherapy using a 1.5 Tesla MR-linac is feasible. A full online workflow adapting to anatomy at each fraction is achievable in approximately 30 min. Intra-fraction bladder filling did not compromise target coverage with the class solution employed.

Current Status of Anatomical Magnetic Resonance Imaging in Brachytherapy and External Beam Radiotherapy Planning and Delivery

Radiotherapy planning and delivery have dramatically improved in recent times. Imaging is key to a successful three-dimensional and increasingly four-dimensional based pathway with computed tomography embedded as the backbone modality. Computed tomography has significant limitations for many tumour sites where soft-tissue discrimination is suboptimal, and where magnetic resonance imaging (MRI) has largely superseded in the diagnostic arena. MRI is increasingly used together with computed tomography in the radiotherapy planning pathway and is now established as a prerequisite for several tumours. With the advent of combined MRI and linear accelerator (MR-linac) systems, a transition to MRI-based radiotherapy planning is becoming reality, with increasing experience and research involving these new platforms. In this overview, we aim to highlight how magnetic resonance-guided imaging has improved radiotherapy, using gynaecological malignancies to illustrate, in both external beam radiotherapy and image-guided brachytherapy, and will assess the early evidence for magnetic resonance-guided radiotherapy using combined MR-linac systems.

Decitabine (DAC) in combination with doxorubicin (DOX) and cyclophosphamide (CTX) in relapsed neuroblastoma (NBL): A Children's Oncology Group Study

9565

Background: Loss of caspase-8 expression causes resistance to apoptosis-inducing agents and potentiates metastasis in NBL. In vitro treatment of NBL cell lines with DAC results in increased caspase-8 expression and enhanced apoptosis following exposure to DOX. As a component of a pediatric phase 1 study, we evaluated low-dose DAC with DOX and CTX in a cohort of children with relapsed NBL. Promoter specific DNA methylation in peripheral blood mononuclear cells (PBMC) was studied. Methods: Patients received a 1-hour infusion of DAC, 5 mg/m2, daily on days 0–6, followed on day 7 by 45 mg/m2 of DOX (with dexrazoxane) and 1 g/m2 of CTX. All patients received G-CSF. Cycles were repeated every 28 days. Methylation-specific and quantitative PCR were used to measure promoter-specific DNA methylation in PBMC prior to and after DAC (day 7 and day 28). Results: Twelve patients with recurrent NBL (median age 9 years, 10 males) were enrolled. DAC/DOX/CTX was well tolerated with grade 4 myelosuppression being the predominant toxicity. Grade 4 neutropenia, median 11 days duration, occurred in 10 pts and grade 4 thrombocytopenia, median 10 days, in 7 patients. Eight patients had progressive disease after = 2 courses of therapy. The remaining 4 patients had stable disease with 5 to 8 courses. Methylation and subsequent demethylation (40–90% decreased methylation) of the MAGE-1 promoter was detected before and after DAC respectively, in PBMC from 6 of 9 patients analyzed to date. Re-expression of MAGE-1 mRNA was also demonstrated in post treatment samples. Caspase-8 demethylation was not detected in PBMC; nor was hypermethylation of caspase-8 detected in pre-treatment PBMC, possibly due to low tumor burden in blood. Conclusions: DAC in combination with cytotoxic chemotherapy was relatively well tolerated in heavily pre-treated children with relapsed NBL. Low-dose DAC induces demethylation and re-expression of MAGE-1 in PBMC, and may be a potential surrogate marker of demethylation within tumor cells. The use of low dose DAC in combination with other agents warrants further study in NBL.

No significant financial relationships to disclose.

Sulfur dioxide inhibition of photosynthesis in isolated spinach chloroplasts

Photosynthetic oxygen evolution by isolated spinach (Spinacia oleracea L.) chloroplasts approached complete inhibition in the presence of a 5 mm concentration of sulfur dioxide. A similar inhibition was observed in the presence of equimolar concentrations of bisulfite ions, suggesting a parallel mode of action. In contrast, an equimolar concentration of sulfite ions was markedly less inhibitory and sulfate ions caused negligible inhibition of apparent photosynthesis. The mode of action of sulfur dioxide and related sulfur anions in inhibiting photosynthesis was found to be essentially independent of direct hydrogen-ion effects. Supplements of inorganic pyrophosphate lessened the inhibition of oxygen evolution caused by sulfur dioxide and the sulfur anions.Sulfur dioxide and the sulfur anions were almost equally effective in inhibiting cyclic and noncyclic photophosphorylation in chloroplast suspensions. However, the extent of the inhibition of these photosynthetic reactions does not appear sufficient to account for the inhibition of photosynthetic oxygen evolution by sulfur dioxide.