Computer Tomography-Based Target Volume Delineation and Clinical Outcomes of Low-Dose Radiation Therapy for Knee Osteoarthritis Authors: Yueh-Feng Lu, Eva Yu-Hsuan Chuang, Zhao-Yu Hsieh, Kai-Xuan Lim, Pei-Wei Shueng, Chen-Xiong Hsu, Deng-Yu Kuo, Pei-Yu Hou, Chih-Hung Chang, Chen-Hsi Hsieh Published: 2026 ABSTRACT PurposeLow-dose radiation therapy is a noninvasive and cost-effective treatment modality used to alleviate pain and improve joint function in patients with knee osteoarthritis. However, standardized guidelines for computed tomography (CT)-based target delineation remain limited. This study aims to evaluate the feasibility and safety of low-dose radiation therapy using a CT-based, anatomy-guided clinical target volume that incorporates both bony and soft tissue structures.Methods and MaterialsWe conducted a retrospective analysis of patients with knee osteoarthritis who received low-dose radiation therapy at our institution. The clinical target volume was delineated on simulation CT to encompass the entire knee joint. Patients received a total dose of 3 Gy, delivered in 6 fractions of 0.5 Gy each on nonconsecutive days. Pain was assessed using a 10-point visual analog scale (VAS) before treatment initiation and upon completion.ResultsA total of 37 patients (median age 69 years; 17 men and 20 women) received treatment across 71 knee sites. The median follow-up was 4.2 months. The median pretreatment VAS score was 6 (range, 4-10), significantly decreasing to 2 (range, 0-5) posttreatment (P < .001). At a minimum follow-up of 2 months, no acute toxicities of grade ≥2 were observed. In a representative case, CT-based delineation enabled inclusion of suprapatellar recess effusion extending beyond conventional field margins, illustrating the potential for improved anatomical coverage.ConclusionsCT-based, anatomy-guided clinical target volume delineation for low-dose radiation therapy in knee osteoarthritis is feasible and safe, with significant pain reduction observed. This approach provides a reproducible framework that may better reflect patient-specific disease extent compared with conventional field-based techniques. Further prospective studies are warranted to validate its clinical impact. Introduction Osteoarthritis (OA) is the most common degenerative joint disorder, affecting more than 500 million people worldwide. It is characterized by progressive loss of cartilage integrity and failure of the synovial joint.1 The knee joint is the most commonly affected site, particularly among older adults and individuals with obesity, leading to significant pain, stiffness, and loss of range of motion over time. Osteoarthritis not only impairs quality of life but also contributes to an increased healthcare burden globally.2 According to current guidelines from the American College of Rheumatology, treatment options for knee OA include nonpharmacologic therapies (eg, weight management, tibiofemoral bracing), pharmacologic therapies (eg, oral and topical nonsteroidal anti-inflammatory drugs (NSAIDs) or opioids, intra-articular corticosteroids or hyaluronic acid injections), and surgical interventions (eg, total knee arthroplasty) for advanced cases.3 However, for patients who are unsuitable for NSAIDs or fail to respond to conventional treatments, a therapeutic gap remains before invasive procedures can be considered. Low-dose radiation therapy (LDRT) has been commonly used in Europe, particularly in Germany, to treat nonmalignant conditions such as osteoarthritis. The recommended dose ranges from 0.5 to 1.0 Gy per fraction, for a total of 3.0 to 6.0 Gy per course.4 A large multicenter retrospective analysis involving elderly OA patients (aged ≥65 years) demonstrated significant pain improvement in approximately two-thirds of the population studied.5 Although the precise mechanism remains under investigation, both in vitro and in vivo studies have shown a significant reduction in adhesion molecules and proinflammatory cytokines following LDRT, suggesting an enhanced anti-inflammatory effect.6, 7, 8 An interim analysis of the IMMO-LDRT01 study reported a statistically significant difference in the immunophenotypes of peripheral blood cells, as well as pain relief, 6 months after receiving LDRT.7 Increasing clinical evidence supports a satisfactory response in knee OA patients treated with LDRT. An ongoing multi-institutional, sham-controlled randomized trial in South Korea is assessing the Western Ontario and McMaster Universities Osteoarthritis Index subscales of pain, stiffness, and physical function, and aims to explore the optimal (and potentially lower) dose of LDRT for knee OA treatment.9 Despite the growing body of evidence supporting LDRT for knee OA, target delineation remains poorly standardized and is largely based on historical 2-dimensional (2D) field arrangements rather than modern imaging techniques. This limitation may contribute to variability in treatment outcomes across studies and complicates interpretation of conflicting clinical trial results. Prospective data on LDRT remain limited and controversial. The ArthroRad trial demonstrated significant pain reduction at 3 months following LDRT,10 whereas a randomized, sham-controlled trial by Mahler et al11 did not show a significant benefit. Notably, these studies did not incorporate standardized CT-based target delineation strategies, and variations in target definition may partially account for the heterogeneity in reported outcomes. In this context, the target volume for LDRT in knee OA was broadly defined, typically extending 8 cm superiorly and inferiorly from the joint space and 3 cm medially and laterally from the femoral and tibial condyles.8 Although such field-based approaches are simple and reproducible, they do not account for patient-specific anatomical variations, including differences in joint morphology, distribution of synovial inflammation, or the presence of joint effusion. With advancements in imaging, particularly CT-based techniques, the accuracy of target delineation and radiation delivery has significantly improved. These innovations enable precise targeting of inflamed synovial tissue while minimizing radiation exposure to surrounding healthy structures and reducing treatment variability. However, existing CT-based approaches remain limited and typically define the target volume based solely on bony anatomy, without adequately incorporating soft tissue components that are central to the pathophysiology of OA. Therefore, our study aims not only to evaluate the clinical outcomes of LDRT in patients with knee OA, but also to address a critical technical gap by proposing a CT-based, anatomy-guided clinical target volume (CTV) definition incorporating both bony and soft tissue structures. This approach seeks to improve reproducibility, enable individualized target delineation, and better reflect patient-specific inflammatory involvement in knee OA. Methods and Materials Study design, patient selection, and treatment details We conducted a retrospective review of patients with symptomatic knee OA who were treated with volumetric modulated arc therapy (VMAT) using linear accelerators at our institution between August 2024 and April 2025. Eligible patients had radiologically and clinically confirmed knee OA, classified as Kellgren–Lawrence grades 1 to 4. Patients with persistent knee pain despite conservative management (eg, analgesics or intra-articular injections) and no prior history of radiation therapy to the affected knee were included. The total prescribed dose was 3 Gy, delivered in 6 fractions of 0.5 Gy per fraction on nonconsecutive days over a 2-week period. Throughout the treatment course, all patients were evaluated weekly by a radiation oncologist to monitor for acute adverse events. Pain assessments were performed at baseline, 2 weeks after treatment completion, and during monthly follow-ups using a 10-point visual analog scale (VAS). Adverse events were evaluated according to the Common Terminology Criteria for Adverse Events version 5.0. Additional patient data, including demographics, body mass index, medical comorbidities, radiographic findings, and pain medication usage, were collected from electronic medical records for correlation with treatment outcomes. The study was approved by the Far Eastern Memorial Hospital Institutional Review Board (IRB number: 114038-E) and was granted a waiver of patient consent. CT-based target delineation All patients underwent a simulation CT scan of the affected knee in the supine position, using a commercial immobilization device to minimize motion. The scan range extended from the mid-femur to the mid-tibia to ensure inclusion of all relevant anatomical structures and potential sites of referred pain. A slice thickness of 2.5 mm was used to facilitate accurate contouring. The CTV for knee LDRT encompasses the entire knee joint, including the articular cavity, synovial membrane, synovial fluid, and periarticular soft tissues suspected to be involved in the inflammatory process. The proximal boundary extends to the superior border of suprapatellar recess, whereas the distal boundary reaches the tibial tuberosity at the insertion of the patellar ligament. Anteriorly, the CTV is limited by the quadriceps tendon, anterior border of patella, and patellar ligament, whereas posteriorly, it extends to the anterior borders of the gastrocnemius muscle, plantaris muscle, popliteus muscle, and soleus muscle. Laterally, the boundary is defined by the medial border of vastus lateralis muscle, lateral patellar retinaculum, iliotibial band, and biceps femoris muscle, extending distally to the lateral aspects of the proximal tibia and fibula. Medially, the CTV is confined by the lateral border of vastus medialis muscle, medial patellar retinaculum, medial collateral ligament, and sartorius muscle (Fig. 1, Table 1). This delineation ensures comprehensive coverage of the knee joint and surrounding structures potentially involved in the pathological process (eg, synovial effusion, osteophytes) while minimizing unnecessary irradiation of adjacent healthy tissues. Planning target volume (PTV) margin selection was based on institutional setup uncertainty analysis. A 1-cm isotropic expansion was applied, consistent with margins estimated using the van Herk formula,12 to ensure adequate target coverage while accounting for setup variability. In routine clinical practice, patient positioning was verified using weekly portal imaging.Figure 1Clinical target volume (CTV) delineation for low-dose radiation therapy in knee osteoarthritis. (A-F) Representative images showing the anatomical boundaries of the CTV.Figure 1 dummy alt textTable 1Clinical target volume definitions for knee osteoarthritis treatmentTable 1 dummy alt textBoundariesKnee jointProximalSuperior border of suprapatellar recessDistalTibial tuberosityMedialLateral border of vastus medialis muscle/medial patellar retinaculum/medial collateral ligament/sartorius muscleLateralMedial border of vastus lateralis muscle/lateral patellar retinaculum/iliotibial band/biceps femoris muscle/lateral side of tibia and fibulaAnteriorQuadriceps tendon/anterior border of patella/patellar ligamentPosteriorAnterior border of gastrocnemius muscle/plantaris muscle/popliteus muscle/soleus muscle Clinical target volume (CTV) delineation for low-dose radiation therapy in knee osteoarthritis. (A-F) Representative images showing the anatomical boundaries of the CTV. Table 1Clinical target volume definitions for knee osteoarthritis treatmentTable 1 dummy alt textBoundariesKnee jointProximalSuperior border of suprapatellar recessDistalTibial tuberosityMedialLateral border of vastus medialis muscle/medial patellar retinaculum/medial collateral ligament/sartorius muscleLateralMedial border of vastus lateralis muscle/lateral patellar retinaculum/iliotibial band/biceps femoris muscle/lateral side of tibia and fibulaAnteriorQuadriceps tendon/anterior border of patella/patellar ligamentPosteriorAnterior border of gastrocnemius muscle/plantaris muscle/popliteus muscle/soleus muscle Clinical target volume definitions for knee osteoarthritis treatment Statistical analysis All clinical and radiation treatment data were collected and entered into an electronic database for analysis. The primary objective of this study was to evaluate changes in VAS pain scores before and after LDRT. Patient characteristics and study outcomes were summarized using descriptive statistics. Categorical variables are presented as frequencies and percentages, whereas continuous variables are expressed as medians with ranges or mean ± standard deviation. A paired t test was used to compare pretreatment and posttreatment VAS scores. All statistical analyses were conducted using R software (version 4.1.2). Two-tailed P values <.05 were considered statistically significant. Results Patient characteristics A total of 37 patients were included in this study, with a median age of 69 years (range, 49-87). The cohort consisted of 17 men and 20 women, with 71 sites treated using LDRT. Among these, 34 patients received treatment for both knees, whereas the remaining 3 patients were treated for only 1 knee. The majority of patients had Kellgren–Lawrence grade 2 or grade 3 OA, as classified by radiological imaging. Four patients received NSAID during LDRT. Details of the general characteristics are shown in Table 2.Table 2Patient characteristicsTable 2 dummy alt textN = 37Age (y, median)69 (49-87)Sex Men Women17 (46%)20 (54%)Body weight (kg, mean ± SD)66.96 ± 11.81BMI (mean ± SD)26.57 ± 4.27Treatment laterality of knee Left knee only Right knee only Bilateral knee2 (5.4%)1 (2.7%)34 (91.9%)Kellgren–Lawrence grade 1 2 3 42 (5.4%)16 (43.3%)18 (48.6%)1 (2.7%)Prior treatments before radiation therapy Systemic analgesics Hyaluronic acid injection Corticosteroid injection21 (56.8%)10 (27%)3 (8.1%)Abbreviations: BMI = body mass index. Table 2Patient characteristicsTable 2 dummy alt textN = 37Age (y, median)69 (49-87)Sex Men Women17 (46%)20 (54%)Body weight (kg, mean ± SD)66.96 ± 11.81BMI (mean ± SD)26.57 ± 4.27Treatment laterality of knee Left knee only Right knee only Bilateral knee2 (5.4%)1 (2.7%)34 (91.9%)Kellgren–Lawrence grade 1 2 3 42 (5.4%)16 (43.3%)18 (48.6%)1 (2.7%)Prior treatments before radiation therapy Systemic analgesics Hyaluronic acid injection Corticosteroid injection21 (56.8%)10 (27%)3 (8.1%)Abbreviations: BMI = body mass index. Patient characteristics Abbreviations: BMI = body mass index. Treatment outcomes All patients completed the planned LDRT treatment of 3 Gy in 6 fractions. The median follow-up was 4.2 months. During posttreatment follow-up, patients experienced a significant improvement in pain scores. The pretreatment median VAS was 6 (range, 4-10, Fig. 2), which significantly decreased to a posttreatment median VAS of 2 (range, 0-5). Statistical analysis using a paired t test demonstrated a significant reduction in pain severity following LDRT (P < .001). At a minimum follow-up of 2 months, no acute grade ≥2 adverse events were observed. Mild and transient erythema (grade 1) was documented in only 1 patient, resolving spontaneously without intervention. Additionally, 1 patient reported slight hyperesthesia, which was self-limiting and resolved within 1 week. No other significant adverse effects were noted during the follow-up period.Figure 2Comparison of pretreatment and posttreatment visual analog scale (VAS) scores. The median pretreatment VAS was 6 (range, 4-10), which significantly decreased to a posttreatment median of 2 (range, 0-5).Figure 2 dummy alt text Comparison of pretreatment and posttreatment visual analog scale (VAS) scores. The median pretreatment VAS was 6 (range, 4-10), which significantly decreased to a posttreatment median of 2 (range, 0-5). A 75-year-old woman presented with significant bilateral knee effusion. Pretreatment CT imaging revealed increased synovial fluid accumulation in the right patellofemoral joint and a prominent popliteal cyst in the left knee. Two months after receiving LDRT, follow-up CT demonstrated a marked reduction in synovial fluid and a noticeable decrease in the size of the popliteal cyst. The estimated synovial fluid volume in the right patellofemoral joint decreased from 43.1 mL to 14.3 mL (Fig. 3A), whereas the volume of the popliteal cyst was reduced from 49.7 mL to 12.1 mL (Fig. 3B).Figure 3Improvement in knee effusion following low-dose radiotherapy. (A) Axial CT images of the right knee showing a reduction in synovial fluid volume in the patellofemoral joint from 43.1 mL pretreatment to 14.3 mL. (B) Axial CT images of the left knee demonstrating a decrease in popliteal cyst volume from 49.7 mL pretreatment to 12.1 mL posttreatment.Abbreviations: CT = computed tomography.Figure 3 dummy alt text Improvement in knee effusion following low-dose radiotherapy. (A) Axial CT images of the right knee showing a reduction in synovial fluid volume in the patellofemoral joint from 43.1 mL pretreatment to 14.3 mL. (B) Axial CT images of the left knee demonstrating a decrease in popliteal cyst volume from 49.7 mL pretreatment to 12.1 mL posttreatment. Abbreviations: CT = computed tomography. Discussion LDRT has been increasingly explored as a promising noninvasive treatment option for knee OA, offering symptom relief with minimal systemic toxicity.8 In our current study, we established a clearly defined, CT-based target delineation protocol that incorporates both bony and soft tissue anatomy of the affected knee. This comprehensive approach resulted in significant pain reduction among treated patients, with a mean decrease of 4.6 points on the VAS score. Notably, a patient presenting with substantial fluid retention in the patellofemoral joint and popliteal region also demonstrated a marked reduction in joint effusion following LDRT. Importantly, our study does not aim to re-establish the clinical efficacy of LDRT, which has already been demonstrated in multiple retrospective and prospective studies. Rather, it focuses on a fundamental yet underexplored issue, the lack of standardized target volume definition in modern radiation therapy. We propose a CT-based, anatomy-guided CTV framework that incorporates both bony and soft tissue structures, enabling individualized target delineation that may better reflect the biological extent of disease. As a weight-bearing synovial joint, the knee has a complex structure comprising articular cartilage, synovial membrane, synovial fluid, and supporting ligaments, all of which contribute to joint function and stability. Pain associated with OA of the knee is multifaceted, involving nociceptive, inflammatory, and neuropathic components. Nerve endings frequently proliferate in the synovium, subchondral bone, and osteochondral junction, contributing to joint hypersensitivity.13,14 Given this pathophysiology, our approach was designed to encompass the entire synovial joint, including the joint capsule and synovial membrane, thereby ensuring coverage of key structures involved in pain generation. The heterogeneity in reported clinical outcomes, including the negative findings from the randomized trial by Mahler et al,11 may partly be explained by differences in target definition and treatment delivery. The lack of standardized, image-guided target delineation could result in suboptimal coverage of inflamed tissues. Previous studies have defined target volumes for knee OA, most were based on 2D or early 3-dimensional planning techniques (Table 3).8,12,13 For instance, Dove et al. adapted guidelines from the Deutsche Gesellschaft für Radioonkologie, defining treatment fields using fixed distances: 8 cm superior and inferior to the joint space, 3 cm medial to the medial femoral condyle, and 3 cm lateral to the lateral tibial condyle.8 A key limitation of conventional field-based techniques lies in their reliance on predefined geometric margins. In our cohort, a representative patient demonstrated suprapatellar recess effusion extending approximately 10.5 cm superior to the knee joint space, exceeding the conventional superior field extent. As illustrated in Fig. 4, this observation suggests that the anatomical extent of inflammatory involvement may surpass uniform geometric definitions and may not be fully encompassed in certain patients. Patient-specific variations in limb length, joint space anatomy, soft tissue distribution, and compartmental OA severity further limit the adequacy of fixed-field approaches and may lead to under-coverage of symptomatic regions or unnecessary irradiation of uninvolved tissues. In addition, without CT-based planning, features such as synovitis, joint effusion, and soft tissue involvement cannot be directly visualized, increasing the risk of geographic miss. Although conventional 2D techniques have demonstrated clinical efficacy, CT-based planning enables visualization of patient-specific anatomical and inflammatory features and allows improved sparing of surrounding normal tissues. This is particularly relevant for posterior muscle groups and uninvolved soft tissues, supporting adherence to the ALARA (As Low As Reasonably Achievable) principle. CT-based approaches also facilitate individualized and reproducible target delineation, which may enhance consistency of treatment delivery.Table 3Comparison of target volume in knee osteoarthritisTable 3 dummy alt textTarget volumePTV marginDoseDove et al.8Superior: 8 cm superior to joint spaceInferior: 8 cm inferior to joint spaceMedial: 3 cm medial to medial femoral condyleLateral: 3 cm lateral to lateral tibial condyleN/A0.5 Gy for 6 fractions every other day or twice a weekSpanish group12PTV:Superior: 2 cm above femoral condylesInferior: 2 cm below tibial condyles and fibula headMedial: 1 cm around of the soft tissues surroundingLateral: 1 cm around of the soft tissues surroundingProximal: 1 cm above lateral and medial femoral epicondyleDistal: Lateral and medial tibial condyles and fibula headN/AN/AAustralian case series13CTV:Physician dependent skin markings for superior and inferior bordersAuto-contouring of bony structures comprised the CTV2 mm0.5 Gy for 6 fractionsevery other day or twice a weekOur studyCTV*:Proximal: Superior border of suprapatellar recessDistal: Tibial tuberosityMedial: Lateral border of vastus medialis muscle/medial patellar retinaculum/medial collateral ligament/sartorius muscleLateral: Medial border of vastus lateralis muscle/lateral patellar retinaculum/iliotibial band/biceps femoris muscle/lateral side of tibia and fibulaAnterior: Quadriceps tendon/anterior border of patella/patellar ligamentPosterior: Anterior border of gastrocnemius muscle/ plantaris muscle/popliteus muscle/ soleus muscle1 cm0.5 Gy for 6 fractionsevery other dayAbbreviations: CTV = clinical target volume; N/A = not applicable; PTV = planning target volume.⁎The knee effusion should be fully included within the CTV.Figure 4Illustrative comparison between conventional field-based coverage and CT-based anatomy-guided target delineation in knee osteoarthritis. (A) Conventional field-based coverage using fixed geometric margins. (B) CT-based anatomy-guided CTV delineation in the coronal view. (C) CT-based anatomy-guided CTV delineation in the sagittal view, demonstrating inclusion of patient-specific soft tissue components, including suprapatellar recess effusion (green), within the defined PTV (red). In this representative case, the effusion extends approximately 10.5 cm superior to the knee joint space, exceeding the conventional superior field extent of 8 cm.Abbreviations: CT = computed tomography; CTV = clinical target volume; PTV = planning target volume.Figure 4 dummy alt text Table 3Comparison of target volume in knee osteoarthritisTable 3 dummy alt textTarget volumePTV marginDoseDove et al.8Superior: 8 cm superior to joint spaceInferior: 8 cm inferior to joint spaceMedial: 3 cm medial to medial femoral condyleLateral: 3 cm lateral to lateral tibial condyleN/A0.5 Gy for 6 fractions every other day or twice a weekSpanish group12PTV:Superior: 2 cm above femoral condylesInferior: 2 cm below tibial condyles and fibula headMedial: 1 cm around of the soft tissues surroundingLateral: 1 cm around of the soft tissues surroundingProximal: 1 cm above lateral and medial femoral epicondyleDistal: Lateral and medial tibial condyles and fibula headN/AN/AAustralian case series13CTV:Physician dependent skin markings for superior and inferior bordersAuto-contouring of bony structures comprised the CTV2 mm0.5 Gy for 6 fractionsevery other day or twice a weekOur studyCTV*:Proximal: Superior border of suprapatellar recessDistal: Tibial tuberosityMedial: Lateral border of vastus medialis muscle/medial patellar retinaculum/medial collateral ligament/sartorius muscleLateral: Medial border of vastus lateralis muscle/lateral patellar retinaculum/iliotibial band/biceps femoris muscle/lateral side of tibia and fibulaAnterior: Quadriceps tendon/anterior border of patella/patellar ligamentPosterior: Anterior border of gastrocnemius muscle/ plantaris muscle/popliteus muscle/ soleus muscle1 cm0.5 Gy for 6 fractionsevery other dayAbbreviations: CTV = clinical target volume; N/A = not applicable; PTV = planning target volume.⁎The knee effusion should be fully included within the CTV. Comparison of target volume in knee osteoarthritis Abbreviations: CTV = clinical target volume; N/A = not applicable; PTV = planning target volume. The knee effusion should be fully included within the CTV. Illustrative comparison between conventional field-based coverage and CT-based anatomy-guided target delineation in knee osteoarthritis. (A) Conventional field-based coverage using fixed geometric margins. (B) CT-based anatomy-guided CTV delineation in the coronal view. (C) CT-based anatomy-guided CTV delineation in the sagittal view, demonstrating inclusion of patient-specific soft tissue components, including suprapatellar recess effusion (green), within the defined PTV (red). In this representative case, the effusion extends approximately 10.5 cm superior to the knee joint space, exceeding the conventional superior field extent of 8 cm. Abbreviations: CT = computed tomography; CTV = clinical target volume; PTV = planning target volume. More recent efforts to incorporate CT-based planning have improved geometric accuracy; however, limitations remain. A Spanish group15 proposed a CT-based method in which the PTV was defined directly: 2 cm above the femoral condyles, 2 cm below the tibial condyles and fibular head, and 1 cm beyond the surrounding soft tissues. Although this approach offers greater precision than traditional 2D methods, it has notable limitations. Chief among them is the absence of a clearly defined CTV, which may result in suboptimal target coverage and increased exposure of normal tissues. An Australian case series was the first to report on the use of VMAT for OA treatment using LDRT.16 In their approach, superior and inferior borders of the CTV were determined by individual physicians, and auto-contouring of bony structures was used to define the CTV. The PTV was generated by applying a 2-mm uniform expansion. However, this method presents limitations. Physician dependent skin marks used to define the proximal and distal boundaries introduce interobserver variability, which may lead to inconsistent target definitions, potentially resulting in either under-coverage of symptomatic regions or unnecessary irradiation. Furthermore, limiting auto-contouring to bony structures overlooks important soft tissue components such as the synovium, cartilage, and joint capsule, which play a central role in the pathophysiology of OA. This could result in incomplete target coverage, especially in cases with active inflammation or effusion. In addition, a narrow 2-mm PTV margin may be insufficient to account for patient positioning uncertainties, increasing the risk of geometric miss and suboptimal dose delivery. In comparison, our approach explicitly incorporates both bony and soft tissue structures, including synovium, joint effusion, and periarticular tissues. This is particularly relevant given that synovial thickening and effusion are frequently observed in patients with symptomatic OA.17 Although a direct comparison with conventional techniques was not performed, CT-based planning implemented with VMAT allows improved conformity to complex target volumes defined on imaging, particularly when incorporating soft tissue structures. The absence of grade 2 or more severe acute adverse events during follow-up further supports the safety of this approach. These findings are consistent with previous studies reporting favorable outcomes using LDRT for OA, suggesting its potential as an adjunctive therapy for pain management with minimal toxicity.18 Given that the total radiation dose used in LDRT is significantly lower than that used in oncologic settings, the theoretical risk of radiation induced malignancies remains minimal.19,20 Despite the promising results, several limitations of this study should be acknowledged. First, the sample size was relatively small. Second, some patients had a short follow-up period, with a minimum duration of 2 months. Although early pain relief was observed, the long-term durability of this effect remains uncertain and requires further observation. Studies with extended follow-up that evaluate pain recurrence, functional outcomes, and disease progression are essential to fully establish the therapeutic role of LDRT in knee OA. Another important limitation is the absence of a control group, which makes it difficult to distinguish the true biological effects of LDRT from possible placebo responses. Given the subjective nature of pain assessment, future randomized controlled trials with sham irradiation arms are necessary to confirm these findings. In addition, although the VAS provides useful insight into pain intensity, incorporating other patient reported outcome measures such as the Western Ontario and McMaster Universities Osteoarthritis Index, the Knee Injury and Osteoarthritis Outcome Score, or the Brief Pain Inventory would offer a more complete evaluation of functional improvement and quality of life. Beyond subjective patient reported outcome measures, future research would greatly benefit from incorporating magnetic resonance imaging. Magnetic resonance imaging Osteoarthritis Keen Score can provide invaluable insights into crucial pathological features of knee OA, such as bone marrow lesions and subchondral cysts, which are increasingly recognized as important biomarkers for disease activity and progression. Visualizing changes in these structural elements would offer a more complete understanding of its potential disease-modifying effects beyond mere symptomatic relief, providing a more robust measure of treatment success. In conclusion, our study demonstrates the feasibility and safety of CT-based, anatomy-guided target delineation incorporating both bony and soft tissue structures for LDRT in knee OA. This approach represents a shift from traditional field-based techniques toward individualized, image-guided radiation therapy planning. Standardizing such contouring strategies may improve reproducibility and provide a biologically relevant framework for future clinical studies. Disclosures The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.