The FK506-binding protein (FKBP) family plays a key role in a variety of tumors and is involved in the regulation of important signaling pathways including AKT, NF-κB and p53, which affects cell proliferation, migration, and multiple cell death modes. Here, we summarize the findings that different FKBP family members exhibit dual functions of promoting or inhibiting tumorigenesis in different types of tumors. The expression levels of FKBP family members are closely related to the prognosis of patients, thus might be used as potential diagnostic and prognostic biomarkers. In the future, it is necessary to combine single-cell sequencing to resolve the spatial distribution of FKBP isoforms, develop clinical validation to promote the translation from molecular mechanism to precision therapy.

This study was conducted to develop and validate a novel deep reinforcement learning (DRL) algorithm incorporating the segment anything model (SAM) to enhance the accuracy of automatic contouring organs at risk during radiotherapy for cervical cancer patients. CT images were collected from 150 cervical cancer patients treated at our hospital between 2021 and 2023. Among these images, 122 CT images were used as a training set for the algorithm training of the DRL model based on the SAM model, and 28 CT images were used for the test set. The model’s performance was evaluated by comparing its segmentation results with the ground truth (manual contouring) obtained through manual contouring by expert clinicians. The test results were compared with the contouring results of commercial automatic contouring software based on the deep learning (DL) algorithm model. The Dice similarity coefficient (DSC), 95th percentile Hausdorff distance, average symmetric surface distance (ASSD), and relative absolute volume difference (RAVD) were used to quantitatively assess the contouring accuracy from different perspectives, enabling the contouring results to be comprehensively and objectively evaluated. The DRL model outperformed the DL model across all evaluated metrics. DRL achieved higher median DSC values, such as 0.97 versus 0.96 for the left kidney (P < 0.001), and demonstrated better boundary accuracy with lower HD95 values, e.g., 14.30 mm versus 17.24 mm for the rectum (P < 0.001). Moreover, DRL exhibited superior spatial agreement (median ASSD: 1.55 mm vs. 1.80 mm for the rectum, P < 0.001) and volume prediction accuracy (median RAVD: 10.25 vs. 10.64 for the duodenum, P < 0.001). These findings indicate that integrating SAM with RL (reinforcement learning) enhances segmentation accuracy and consistency compared to conventional DL methods. The proposed approach introduces a novel training strategy that improves performance without increasing model complexity, demonstrating its potential applicability in clinical practice.

Personalized precision irradiation of patients with left‐sided breast cancer is possible by examining the setup errors of 3‐ and 4‐mm gated window widths for those treated with deep inspiration breath‐hold (DIBH) treatment. An observational study was performed via a retrospective analysis of 250 cone‐beam computed tomography (CBCT) images of 60 left‐breast cancer patients who underwent whole‐breast radiotherapy with the DIBH technique between January 2021 and 2022 at our hospital. Among them, 30 patients had a gated window width of 3 mm, while the remaining 30 had a gated window width of 4 mm; both groups received radiotherapy using DIBH technology. All patients underwent CBCT scans once a week, and the setup errors in the left‐right (x‐axis), inferior‐superior (y‐axis), and anterior‐posterior (z‐axis) directions were recorded. The clinical‐to‐planning target volume (CTV‐PTV) margins of the two gating windows were calculated using established methods. The setup error in the Y direction was 1.69 ± 1.33 mm for the 3‐mm – wide gated window and 2.42 ± 3.02 mm for the 4‐mm – wide gated window. The two groups had statistically significant differences in the overall mean setup error (Dif 0.7, 95% CI 0.15–1.31, t = 2.48, p= 0.014). The Z‐direction setup error was 2.32 ± 2.12 mm for the 3‐mm – wide gated window and 3.15 ± 3.34 mm for the 4‐mm – wide gated window. The overall mean setup error was statistically significant between the two groups (Dif 0.8, 95% CI 0.13–1.53, t= 2.34, p = 0.020). There was no significant difference in the X‐direction setup error (p > 0.05). Therefore, the CTV‐PTV margin values for a 3‐mm gated window width in the X, Y, and Z directions are 5.51, 5.15, and 7.28 mm, respectively; those for a 4‐mm gated window width in the X, Y, and Z directions are 5.52, 8.16, and 10.21 mm, respectively. The setup errors of the 3‐mm – wide gating window are smaller than those of the 4‐mm – wide gating window in the three dimensions. Therefore, when the patient's respiratory gating window width is reduced, the margin values of CTV‐PTV can be reduced to increase the distance between the PTV and the organs at risk (OARs), which ensures adequate space for the dose to decrease, resulting in lower dose exposure to the OARs (heart, lungs, etc.), thus sparing the OARs from further damage. However, some patients with poor pulmonary function or unstable breathing amplitudes must be treated with a slightly larger gating window. Therefore, this study lays a theoretical basis for personalized precision radiotherapy, which can save time and reduce manpower in the delivery of clinical treatment to a certain extent. Another potential benefit of this work is to bring awareness to the potential implications of a slightly larger gating window during treatment without considering the resulting dosimetric impact.

This study evaluates the differences in application of the RayStation Monte Carlo algorithm (RMC) compared to Acuros XB (AXB) and the Anisotropic Analytical Algorithm (AAA) in Eclipse for conventional radiotherapy planning of lung cancer using the novel ring-shaped Halcyon accelerator. A total of 63 non-small-cell lung cancer patients were retrospectively included, with a prescription dose of 60 Gy delivered in 30 fractions. Radiotherapy plans were initially designed and optimized in RayStation, then recalculated in Eclipse using AXB and AAA to assess algorithmic differences in dose distributions. Analysis of data from 63 patients, combined with simulations using square fields and cylindrical water phantoms, revealed that RMC and AXB achieved high consistency in target dose coverage and conformity. High-dose indicators, such as D_2% and D_0.03cc, showed close agreement between RMC and AAA, while AXB tended to slightly underestimate peak doses. For prescription dose coverage metrics like D_95% and D_98%, the difference between RMC and AXB was less than 1%, whereas AAA exhibited minor degradation. In organ-at-risk dose evaluations, RMC delivered higher doses compared to AXB and AAA, with AAA doses exceeding those of AXB. These findings confirm the dose consistency of RayStation and Eclipse algorithms for use with the Halcyon accelerator. The RayStation Monte Carlo algorithm (RMC) is a viable alternative to AXB, especially in lung cancer cases with high tissue heterogeneity, as target coverage and conformity discrepancies remain within 0.5%. Additionally, recalculation of RMC-optimized plans in Eclipse using AXB resulted in lower organ-at-risk doses. Therefore, recalculations performed before treatment do not compromise the clinical adequacy of volumetric dose evaluations.

Objective: To explore the advantages of dosimetry and the treatment efficiency of tangent-arc technology in deep inspiration breath-hold radiotherapy for breast cancer.Methods: Forty patients with left-sided breast cancer who were treated in our hospital from May 2020 to June 2021 were randomly selected and divided into two groups. The first group’s plan was a continuous semi-arc that started at 145° ( ± 5°) and stopped at 325° ( ± 5°). The other group’s plan, defined as the tangent-arc plan, had two arcs: the first arc started at 145° ( ± 5°) and stopped at 85° ( ± 5°), and the second arc started at 25° ( ± 5°) and stopped at 325° ( ± 5°). We compared the target dose, dose in organs at risk (OARs), and treatment time between the two groups.Results: The target dose was similar between the continuous semiarc and tangent-arc groups. The V₅ of the right lung was significantly different between the two groups (Dif 5.52, 95% confidence interval 1.92-9.13, t=3.10, P=0.004), with the patients in the continuous semi-arc and tangent-arc groups having lung V₅ values of (9.16 ± 1.62)%, and (3.64 ± 0.73)%, respectively. The maximum dose to the spinal cord was (1835.88 ± 222.17) cGy in the continuous semi-arc group and (599.42 ± 153.91) cGy in the tangent-arc group, yielding a significant difference between the two groups (Dif 1236.46, 95% confidence interval 689.32-1783.6, t=4.57, P<0.001). The treatment times was (311.70 ± 60.45) s for patients in the continuous semi-arc group and (254.66 ± 40.73) s for patients in the tangent-arc group, and there was a significant difference in the mean number of treatment times between the two groups (Dif 57.04, 95% confidence interval 24.05-90.03, t=3.5, P=0.001).Conclusion: Both the continuous semi-arc and tangent-arc plans met the clinical prescription dose requirements. The OARs received less radiation with the tangent-arc plan than the continuous semi-arc plan, especially for the lung (measured as V₅) and the spinal cord (measured as the maximum dose). Tangent-arc plan took significantly less time than the continuous semi-arc, which can greatly improve treatment efficiency. Therefore, tangent-arc plans are superior continuous semi-arc plans for all cases.

Background: The use of the deep inspiration breath-hold (DIBH) technique reduces cardiac and lung radiation exposure during left breast cancer radiotherapy. However, the optimal beam delivery technique and the effects of patient adaptation during DIBH remain incompletely understood. Objective: In this study, the dosimetric differences between continuous semi-arc and tangent-arc plans in stage I left-sided breast cancer patients using DIBH were compared, and the treatment session duration was descriptively analyzed to characterize treatment-time trends during routine DIBH delivery. Methods: Twenty patients treated at our hospital from 01/05/2022–31/05/2023 were retrospectively selected from the institutional database for exploratory dosimetric analysis. Two radiotherapy plans were created on the basis of each patient’s computed tomography (CT) images. Dosimetric parameters for the planning target volume (PTV) and organs at risk (OARs), and beam-on and total treatment times, were compared. Results: The conformity index (CI) for the PTV was significantly better with the continuous semi-arc plan ( P < 0.05), whereas the other PTV parameters did not significantly differ between the plans ( P > 0.05). The doses and beam-on time for all OARs (except the left ventricle) were significantly lower for the tangent-arc plan ( P < 0.05). Treatment time tended to stabilize across fractions, with a significant difference between the 15th and 16th sessions ( P < 0.05). Conclusion: With the tangent-arc plan, the beam-on time and radiation exposure to OARs were observed to be lower, while adequate PTV coverage was maintained in patients with stage I left-sided breast cancer using DIBH. Treatment times tended to stabilize with increasing treatment fractions. This observation suggests gradual patient adaptation during routine DIBH rather than a predefined training effect. Given the exploratory nature of these findings and the limited sample size from a single institution, these findings should be interpreted with caution and warrant further investigation in larger, multicenter studies.