Reports of Practical Oncology and Radiotherapy最新文献

Background: Rhabdomyosarcoma (RMS) of the extremities has a particularly poor prognosis compared to other primary sites due to an increased rate of alveolar histology, higher rate of metastasis, and the extent of regional lymph node involvement. To date there are few assessments comparing upper extremity (UE) to lower extremity (LE) RMS of the extremities using population-based registry, so we sought to compare survival between UE and LE RMS.

Materials and methods: Using the Surveillance, Epidemiology, and End Results (SEER) database, cases of RMS of the UE and LE diagnosed between 2000-2020 were collected. Descriptive statistics and chi-square analyses were completed for one-and five-year survival. Log-rank and Cox regression analyses were completed to compare UE versus LE survival.

Results: A total of 641 cases were included, of which 221 (34.5%) were UE and 420 (65.5%) were LE. On log-rank tests, UE survival was longer than LE survival (p = 0.021). The one-year survival rate was greater for the UE (88.7%) compared to the LE (81.4%) (p = 0.020) but similar at five-years. Cox regression analysis showed no difference in survival between UE and LE primary site (hazard ratio = 1.172, p = 0.322).

Conclusions: In comparing UE and LE RMS survival, UE survival was greater at one-year, but not on adjusted analyses. These findings contribute to the few prior assessments of outcomes between UE and LE RMS, though direct comparisons between UE and LE should be included in future prospective studies.

Background: The main treatment method for head and neck squamous cell carcinomas (HNSCCs) is surgery in combination with radiotherapy or chemoradiation. However, the tumor heterogeneity and tumor microenvironment are issues of radiotherapy success. Due to this fact, the radioresistance process is not fully understood and seems to be a challenge for current oncology.

Materials and methods: Radiotherapy treated HNSCC patients were divided into two groups based on the overall survival and excluding those with human papillomavirus (HPV) infection or treated with chemotherapy or targeted therapy. Next, groups were compared based on the clinical-pathological and transcriptome data (RNAseq) from The Cancer Genome Atlas Project (TCGA) using Gene Set Enrichment Analysis (GSEA) software and GraphPad Prism toll.

Results: A model was created, consisting of two contrasting groups of patients: effective treatment group (ETG, n = 34) and ineffective treatment group (ITG, n = 31) for radiotherapy. Patients in the ITG group had a significantly shorter progression-free interval (PFI) than the ETG group, with a median of 266 days (p < 0.0001). Between the ETG and ITG groups, no differences (p > 0.05) were observed in clinical and pathological parameters, except perineural invasion (p = 0.0068) and the presence of a new tumor event after initial treatment (p < 0.0001). Molecular pathway analysis showed that ITG patients had statistically significantly increased expression of genes associated with DNA repair.

Conclusions: We observed that our model, consisting of two groups, differed at the molecular level in genetic changes. Moreover, the presented model and its characterization showed that it was potentially useful for searching for potential biomarkers.

Background: Efficient and accurate delineation of target volumes and organs-at-risk (OARs) is critical in radiotherapy planning. Deep learning-based auto-contouring improves consistency and reduces manual effort, but integrating AI tools into established workflows is challenging when platforms differ. This work presents an automated integration of RayStation's AI-based auto-contouring with the Eclipse treatment planning system (TPS).

Materials and methods: Custom scripts were developed in Eclipse (C#) and RayStation (Python), supported by DICOM networking, to enable automated data exchange. The Eclipse script exports computed tomography (CT) images and a list of requested structures to RayStation. Upon receipt, RayStation 2024B automatically triggers a Python script that performs deep learning segmentation using built-in models. The resulting contours are then exported back to Eclipse for review, requiring no user interaction beyond initial input.

Results: The integration was validated on 35 clinical cases. The complete process, from CT export to contour availability in Eclipse, averaged 1.1 minutes per case. The solution significantly reduces contouring time and improves workflow consistency.

Conclusion: This automated cross-platform integration enables deep learning segmentation from RayStation to be accessed directly within Eclipse, offering a streamlined and scalable contouring solution. It enhances planning efficiency, reduces inter-observer variability, and allows clinicians to benefit from state-of-the-art AI tools within their existing TPS. The framework is modular and adaptable to evolving clinical protocols, serving as a practical example of multi-vendor system interoperability in modern radiotherapy planning.

Background: Pulmonary toxicity can occur during radiation therapy of the lungs. Dose metrics evaluated at the lobar level can improve the ability to predict toxicity. Contouring lung lobes is challenging and time consuming. Currently there are limited dosimetry studies evaluating the dose to lung lobes. The purpose of this work was to develop and validate an artificial intelligence (AI) lung lobe auto-contouring algorithm using radiation therapy planning images.

Materials and methods: Fifty lung cancer patients from two institutions were analyzed, and a clinician contoured all five lung lobes [left upper lobe (LUL), left lower lobe (LLL), right upper lobe (RUL), right middle lobe (RML), and right lower lobe (RLL)] on the free-breathing computed tomography data set. The AI model used a residual 3D U-Net and trained using the expert lobe contours for forty patients. Validation was carried out by comparing expert lobe contours on ten patients against AI-based lobe contours using dice similarity coefficients (DSC).

Results: The AI-based model showed good agreement with expert contours with overall DSC of 0.93 (range of 0.78-0.97). The DSC were 0.95 (0.97-0.91), 0.92 (0.96-0.85), 0.94 (0.97-0.87), 0.88 (0.93-0.78), and 0.94 (0.96-0.91), for the LUL, LLL, RUL, RML, and RLL, respectively.

Conclusions: This work presents a validation of AI-based lung lobe contours on free-breathing data and shows good agreement with expert contours.

Background: Intraoperative radiation therapy (IORT) delivered during breast-conserving surgery may streamline treatment and enhance cosmetic outcomes. However, optimal patient selection criteria remain uncertain, especially in Asian populations, where international guidelines may not be fully applicable. This study aimed to evaluate local recurrence rates, identify key prognostic factors, and assess the impact of supplemental external beam radiation therapy (EBRT) in Taiwanese patients receiving IORT.

Materials and methods: This retrospective analysis included 1,306 Taiwanese patients who underwent IORT between June 2014 and December 2021. Patients were stratified into low-risk and high-risk groups based on institutional criteria. Low-risk patients were those meeting all of the following: age ≥ 45 years, tumor size ≤ 3.5 cm, negative nodal involvement, Ki-67 ≤ 30%, hormone receptor (HR) positivity, and HER2 negativity. Patients who did not meet one or more of these criteria were classified as high-risk and were recommended to receive supplemental external beam radiation therapy (EBRT) following IORT.

Results: In our cohort of 1,306 patients, the 3-year disease-free survival rate was 93.11% [95% confidence interval (CI): 91.6-94.4%], with 87 patients (6.89%) experiencing recurrence. The median age was 52 years, and the median follow-up duration was 30.5 months. The overall local recurrence rate was 9.3%, which declined to 1.78% after excluding patients who did not receive indicated adjuvant therapies. Multivariable analysis identified Ki-67 > 30%, omission of hormone therapy, and omission of recommended EBRT as significant predictors of local recurrence, whereas age < 45 was not independently associated with increased risk. Among high-risk patients, the addition of EBRT was associated with a 48.8% reduction in recurrence, achieving outcomes comparable to those of low-risk patients.

Conclusions: Age alone should not determine IORT eligibility. A multifaceted approach, including tumour biology and adherence to recommended therapies, is essential. Supplemental EBRT improves outcomes in high-risk patients, and adapting guidelines may enhance patient selection in Asian populations.

Background: The effectiveness of external beam photon radiotherapy relies on precise planning and execution, involving patient-specific quality assurance to ensure clinical goals are met. This involves verifying the therapy plan against constraints set by the oncology team and confirming its correct implementation. Standard QA methods include gamma analysis and dose-volume histograms (DVH), each with limitations. Gamma analysis highlights delivery errors without distinguishing between over- and under-dosage, while DVH can show coinciding planned and delivered doses despite spatial distribution differences. In this work, a novel method, quantitative isodose analysis, is evaluated on treatment plans with the purpose of determining whether it can address these limitations, and assessing their clinical significance.

Materials and methods: Quantitative isodose analysis works by comparing planned and delivered isodoses for each structure, identifying volumes of under- and over-dosage. This method was validated using anonymized patient data involving 75 treatment plans for different cancer locations.

Results: In a prostate cancer case, significant rectal overdosage was detected, leading to a treatment plan revision. Another rectal cancer case showed underdosage but met clinical implementation criteria.

Conclusions: Results showed that quantitative isodose analysis could detect clinically significant dose delivery errors that standard QA might miss. The findings suggest that while standard QA methods are essential, quantitative isodose analysis provides an additional layer of scrutiny, enhancing the detection of dose discrepancies. The study highlights the potential of quantitative isodose analysis in improving radiotherapy QA, despite limitations such as reliance on reconstructed 3D dose distributions and the need for broader patient cohort studies.

Background: A significant minority of patients in high income countries with symptomatic mucosal head and neck squamous cell carcinomas (HNSCC) warrant treatment of their locoregional disease but are not suitable for standard high dose radiation therapy (RT) with or without concurrent chemotherapy. This study aimed to determine the factors associated with locoregional control (LRC) and survival for patients undergoing high dose palliative-intent RT, to help improve patient selection for this treatment approach.

Materials and methods: This retrospective cohort study included all patients with HNSCC who received high dose RT (50-55 Gy in 20 fractions over 4 weeks) with palliative-intent from 2007-2024 at an academic Australian cancer centre.

Results: 53 patients comprised the study cohort, of which 92% completed the prescribed RT in full. Median overall survival was 21.6 months and in-field LRC at 12-months was 80%. Acute toxicities were low [Common Terminology Criteria for Adverse Events (CTCAE) grade 3 mucositis 17%, local pain 6.5%, and dysphagia 4.4%, with no grade 4-5 toxicities], with resolution of majority by six months post RT (7% grade 2, no grade 3 or higher toxicities). Larger primary tumours (T3 or T4) and more advanced stage disease [American Joint Committee on Cancer (AJCC) stage III-IV, 8^th edition] were associated with worse in-field LRC.

Conclusions: High dose palliative-RT in patients with mucosal HNSCC not suitable for definitive chemoradiotherapy provided durable local control with low toxicities after 12 months. In-field locoregional failures were more likely for more advanced disease.

Background: This research assesses the dosimetric differences in volumetric modulated arc therapy (VMAT) between True- Beam^TM C-arm and Halcyon™ ring gantry linacs, specifically for craniospinal irradiation (CSI) treatment.

Materials and methods: The fifteen patients retrospectively chosen for this study were treated with the TrueBeam linear accelerator between 2020 and 2023. Initially, VMAT patient plans were planned and delivered in TrueBeam (TB6X), and using the same objective function, TB6FFF and Halcyon™ (HAL6FFF) plans were generated. Plans were compared dosimetrically using dose volume histogram (DVH) parameters for the cranial, spine targets and Organs At Risk (OARs) in the treatment planning system (TPS). The different dose dosimetric criteria were evaluated such as dose received by D_98%, D_95%, D_50%, D 1cc, volume receiving ≥ 107% dose, max dose (Gy), homogeneity index (HI), conformity index (CI), Monitor Units (MU) and dose spillage [V_20Gy, V_10Gy, V_5Gy, V_2Gy (%), D_Body - planning target volume (PTV)]. Delivery accuracy was evaluated by performing pre-treatment Portal Dosimetry (PD) measurements applied with 3%/2 mm acceptance criteria and a 95% passing rate.

Results: Brain PTV dose received by 95% of the PTV volume (D_95% (Gy) PTV) for TB6X, TB6FFF, and HAL6FFF plans were 35.9 ± 0.3 Gy, 35.7 ± 0.4 Gy, and 35.6 ± 0.4 Gy respectively and the difference was statistically insignificant (paired sample t-test). Spine PTV dose D_95% in the same sequence, 35.7 ± 0.3 Gy, 35.5 ± 0.6 Gy, and 35.4 ± 0.3 Gy with a statistically significant difference (p = 0.016) for TB6X vs. HAL6FFF. Brain PTV HI for TB6X, TB6FFF, and HAL6FFF were 0.07 ± 0.03, 0.08 ± 0.03, and 0.09 ± 0.02, respectively, and in the same sequence, Spine PTV 0.07 ± 0.01, 0.08 ± 0.02 and 0.09 ± 0.02. MUs for TB6X, TB6FFF, and HAL6FFF plans were 1004.9 ± 97.7, 1454.3 ± 197.9, and 1195.8 ± 204.7, respectively, with a mean difference of 449.4, 190.9 and 258.5 MUs and found statistically significant (p < 0.005).

Conclusions: Clinically and dosimetrically acceptable VMAT-based craniospinal irradiation plans can be generated by the newly introduced dual-layer Multi-Leaf Collimator (MLC) model in a Halcyon™ linac.

Background: Histopathological characterization is a common method employed to identify actinic keratosis and squamous cell carcinoma. Oftentimes, formalin-fixed paraffin-embedded (FFPE) skin biopsies are accessible in repositories for retrospective analysis.

Materials and methods: In this study, we extracted total ribonucleic acid (RNA) from FFPE specimens, successfully conducted small RNA sequencing and performed differential expression analysis to assess candidate biomolecular signatures that are unique and similar between different skin lesions and their malignant transformations.

Results and conclusions: Although RNA sequencing of FFPE specimens is challenging and requires further optimization, we identified several candidate microRNAs (miRNAs) across all 3 stages of keratinocytic intraepidermal neoplasia, including the upregulation of hsa-miR-145-5p, hsa-miR-30-5p, and hsa-miR-205-5p and downregulation of hsa-miR-486-5p against healthy controls. These miRNAs may serve as a potential diagnostic, prognostic, and therapeutic biomarkers for skin cancer. Future studies are needed to improve methods for FFPE analysis and to validate the key differences in miRNA signatures involved in skin cancer development.