Dosimetric and Radiobiological Analysis of VMAT Treatment Plan with Flattened and Flattening Filter-free Photon Beams for Postoperative Oral Cavity Cancer Treatment

Introduction

Methods

Patient Selection and CT Simulation
This retrospective analysis involved a cohort of 20 patients diagnosed with oral cavity cancer, sourced from our institutional database. The study focused on histopathologically confirmed cases of individuals aged between 18 and 70 years, all possessing a Karnofsky Performance Status exceeding 70. Patients were excluded if they had residual disease, metastatic conditions, prior radiotherapy exposure, or uncontrolled medical issues.

CT simulation was performed using a Revolution EVO (GE Healthcare) system, with patients positioned in a supine and immobilized with a thermoplastic mask. A standard and contrast-enhanced computed tomography (CT) scan was performed with a slice thickness of 2.5 mm, encompassing the area from the top of the skull to the mid-thoracic region. The obtained CT data were then transferred to the treatment planning system for comprehensive volumetric delineation.

The contouring process meticulously defined the Gross Tumor Volume (GTV), Clinical Target Volume (CTV), and Planning Target Volume (PTV) in accordance with institutional protocols. Critical organs at risk (OARs), including the spinal cord, brainstem, parotid glands, mandible, and structures within the oral cavity, were accurately contoured to ensure precise treatment planning and minimize the risk of radiation- induced complications.

Treatment Planning
This retrospective investigation involved a thorough treatment planning methodology for patients diagnosed with oral cavity cancer. A total of 40 treatment plans were formulated for 20 individuals, utilizing both Flattened Filter (FF) and Flattening Filter Free (FFF) photon beam techniques within the framework of Volumetric Modulated Arc Therapy (VMAT).

Multiple dose levels were allocated to the Planning Target Volumes (PTVs) utilizing the Simultaneous Integrated Boost (SIB) methodology. The dosage for PTV60, which included the high-risk Clinical Target Volume (CTV) and the nodal CTV, was established at 60Gy in 30 fractions @ 2 Gy per fraction. In contrast, the dosage for PTV54, which addressed the low-risk CTV, was determined to be 54 Gy over 30 fractions @ 1.8 Gy per fraction.

The treatment planning process was carried out utilizing the Eclipse Treatment Planning System (version 16.1), with final dose calculations performed through the Anisotropic Analytical Algorithm (AAA). The beam characteristics were clearly specified, with a dose rate of 600 MU/min for flattened fields (FF) and 1400 MU/min for flattening filter-free (FFF) beams. All treatment plans were developed and executed using a True Beam SVC linear accelerator (Varian Medical Systems, a subsidiary of Siemens Healthineers) equipped with a 120 millennium multileaf collimator (MLC). The VMAT treatment planning approach utilized a sophisticated arc rotation configuration. Three distinct arcs of rotation were meticulously crafted, encompassing angles of 181°-179°, 179°-181°, and 181°-179°, with collimator angles precisely established at 30°, 330°, and 30°, respectively. To minimize potential bias and ensure methodological consistency, all planning and optimization parameters were uniformly applied across both FF and FFF photon beam plans. The plan optimization objective was 95% of the PTVs received the prescribed dose, while concurrently reducing exposure to the Organs at Risk (OARs) including Spinal cord: ≤45 Gy, Brainstem: Dmax≤54 Gy, Optic chiasm: Dmax≤54 Gy, Eyes: Dmax≤45 Gy, Parotid glands: Dmean≤26 Gy, D50%≤30 Gy, V30Gy≤50%, V40Gy≤30%, V50Gy≤20%.

The VMAT plans generated by VMAT_FF and VMAT_FFF can be accessed through the integrated dose-volume analysis tool available in the Eclipse system. Furthermore, the Normal Tissue Complication Probability (NTCP) value for Organs at Risk (OARs) can be evaluated using a biological assessment tool, which is not included as a standard feature but is instead an additional software component created by RaySearch Laboratories. Additionally, the NTCP for each OAR was determined utilizing the Poisson model, relying on the parameters and endpoints specified in Table 1.

Table 1: The biological functions selected for the biological evaluation of the oral cavity patient plans

Treatment Plan Evaluation
In the assessment of radiotherapy treatment plans, various dosimetric indices play a vital role in evaluating the quality and efficacy of dose distribution within the target volume, as well as the protection of organs-atrisk (OARs).

Coverage Index: It is defined as ratio of minimum dose within target volume to prescribed dose.[]

C=Dmin / PD

The prescribed dose (PD) represents the minimum dosage required for tumor volume. A treatment plan is deemed compliant with protocol if the target volume is entirely encompassed by 90% of the prescribed isodose. A minor deviation occurs when the target is covered by 80% of the prescribed dose. Conversely, if 80% of the PD fails to fully cover the target, it is classified as a major deviation.[]

Uniformity Index: UI = D5% / D95%,

D5% and D95% denote the doses administered to the highest and lowest 5% of the target volume, respectively.[] A reduced UI signifies an improved dose distribution throughout the target volume. The categorization of UI values is as follows: Excellent for UI≤0.95, Moderate for 0.951.0.

Homogeneity Index: HI=Dmax / PD,

Where, Dmax is the maximum dose delivered to the target and PD is the prescribed dose.[] A reduced HI value signifies improved dose uniformity within the target area. The categorization of HI values is as follows: Excellent for HI≤1, Moderate for 1.11.5.

Conformity Index: CI=Vreference volume/PTV Volume

Where, Vreference volume is the volume receiving the reference dose, and PTV volume is the planning target volume. The theoretical ideal for the CI is 1.[] A CI value falling between 1 and 2 indicates that the treatment aligns with the prescribed treatment plan.

The dose gradient index (GI) assesses the efficacy of dose reduction beyond the planning target volume (PTV). It is determined by the ratio of the volume that receives the prescribed isodose line to the volume that receives half of that dose, allowing for the comparison of treatment plans that exhibit similar dose conformity yet differ in their gradients.[]

The unified dosimetry index (UDI) serves as a thorough assessment tool for radiotherapy treatment plans by integrating key dosimetric factors: Coverage (C), Conformity Index (CI), Homogeneity Index (HI), and Gradient Index (GI). The calculation is expressed as follows: UDI = C×CI×HI×GI

A UDI value approaching 1 signifies an optimal quality of the treatment plan, whereas values exceeding 1 are generally deemed unsatisfactory. This index plays a crucial role in determining the most effective techniques for treatment planning.[]

Gamma Analysis
Patient-specific quality assurance (PSQA) plans were created for 20 treatment plans, with each techniqueby performing calculations on a CT scan of the ArcCHECK phantom within the Eclipse Treatment Planning System (Varian Medical Systems, Palo Alto, CA) and subsequently recalculating doses based on the phantom"s geometry. These PSQA plans were then exported as DICOM files to facilitate patient-specific QA measurements. The measurements utilized a Sun Nuclear ArcCHECK phantom (Sun Nuclear Corporation, Melbourne, FL), which is equipped with 1386 Sun Point diode detectors. All patient-specific QA treatment plans were administered using a Varian True Beam SVC linear accelerator, and the measurements were analyzed with the SNC ArcCHECK and Patient software program. A quantitative assessment of the measured doses in comparison to the doses calculated by the treatment planning system (TPS) was conducted through gamma index analysis. The parameters for absolute gamma analysis were established at a 3 mm distance-to-agreement, a 3% dose difference, and a 10% dose threshold. Gamma passing rates (GPR) of 95% or higher were acceptable.

Statistical Analysis
Statistical analyses were conducted utilizing Jamovi software (version 1.6). The mean values along with standard deviation (SD) for the VMAT_FF and VMAT_FFF plans were computed. The Student"s paired t-test was employed to assess the significance of the differences between these two plans. A probability value (p) of ≤0.05 is regarded as indicative of statistically significant differences between the two methods.

Results

Table 2: Comparison of PTV dosimetric parameters between FF and FFF VMAT plans

For the PTV: 60Gy, the minimum dose administered to the PTV for both VMAT_FF (50.0±4.19 Gy) and VMAT_FFF (50.1±4.26 Gy) is nearly the same, with no statistically significant difference (p=0.710). Likewise, the average dose to the PTV is almost identical for both methods, exhibiting minimal variations-VMAT_FF provides 60.1±0.134 Gy while VMAT_FFF offers 60.1±0.119 Gy. This difference is not statistically significant (p=0.123). The maximum dose is also comparable between the two methods (VMAT_FF: 64.3±0.976 Gy, VMAT_FFF: 64.4±1.00 Gy), with no significant difference observed (p=0.537). The D95% values for both VMAT_FF (96.4±0.773%) and VMAT_FFF (96.4±0.867%) are virtually the same, indicating no significant difference (p=0.993). These findings, as presented in Table 1, imply that both techniques are equivalent in delivering the minimum, mean, and maximum doses, as well as in providing similar coverage for 95% of the PTV. For PTV-54Gy, the average dose administered to the PTV is 54.2±0.217 Gy for VMAT_FF and 54.3±0.217 Gy for VMAT_FFF. The disparity between the two methods is statistically significant (p=0.011).

The evaluation of the plan dosimetric parameters reveals that the uniformity index (UI) for VMAT_FF (1.077±0.013) and VMAT_FFF (1.070±0.016) does not exhibit a significant difference (p=0.438). This suggests that both techniques achieve comparable dose uniformity throughout the PTV. The coverage index values for the two techniques are closely aligned, with VMAT_FF measuring 0.834±0.07 and VMAT_FFF at 0.835±0.07, showing no significant difference (p=0.710). Regarding the homogeneity index (HI), both VMAT_FF (1.07±0.016) and VMAT_FFF (1.07±0.017) are nearly the same, with no significant difference (p=0.537). This finding suggests that both techniques provide comparable homogeneity in dose distribution, ensuring an even dose within the PTV. The conformity index (CI) is also very similar for both methods, with VMAT_FF at 0.975±0.017 and VMAT_ FFF at 0.975±0.019, again showing no significant difference (p=0.813). Both techniques exhibit equivalent conformity. In terms of the gradient index (GI), the values for both techniques are comparable (VMAT_ FF: 1.21±0.120, VMAT_FFF: 1.21±0.118), with no significant difference (p=0.769). This indicates that the dose fall-off outside the PTV is similarly pronounced for both techniques. Lastly, the undefined dosimetric index (UDI) for both techniques is nearly identical (VMAT_FF: 1.05±0.015, VMAT_FFF: 1.05±0.014), with no significant difference (p=0.712). This implies that both techniques possess similar dosimetric characteristics that are not reflected in the other indices, thereby maintaining equivalent treatment quality.

The monitor units (MU) were notably higher in the VMAT_FFF plans (583±52.1) compared to the VMAT_FF plans (530±69.9, p=0.001), indicating improved delivery efficiency in the VMAT_FF plans. The mean Gamma Passing Rate was significantly higher for VMAT_FF (98.8±0.359%) compared to VMAT_FFF (98.5±0.346%), p<0.001.

Table 3 summarizes the details of the dosimetric parameters and NTCP values associated with different organs at risk (OARs) in the comparison between flattened field (FF) and flattened filter-free (FFF) VMAT plans. The maximum dose delivered to the spinal cord was identical for both the VMAT_FF (31.8±3.29 Gy) and VMAT_FFF (31.8±3.28 Gy) treatment plans, demonstrating no statistically significant difference (p=0.834). Similarly, the brainstem received a maximum dose of 22.1±10.4 Gy in the VMAT_FF plan and 22.4±11.0 Gy in the VMAT_FFF plan, with no significant variation noted (p=0.582). Furthermore, the maximum dose to the optic chiasm was slightly lower in the VMAT_FFF plan (2.45±1.99 Gy) compared to the VMAT_FF plan (2.52±1.98 Gy), although this difference was not statistically significant (p=0.278).

Table 3: Comparison of OAR physical and biological dose analysis between FF and FFF VMAT plans

Regarding the mandible, no significant difference was observed (p=0.694), with doses of 62.2±0.725 Gy for the VMAT_FF plan and 62.3±0.757 Gy for the VMAT_ FFF plan. For the Right Parotid Gland, the average dose delivered by VMAT_FF is 24.6±4.64 Gy, whereas VMAT_FFF provides a dose of 24.1±4.52 Gy. The observed difference is statistically significant (p=0.001), indicating that VMAT_FF delivers a marginally higher mean dose. Regarding D50, which represents the dose received by 50% of the parotid volume, VMAT_FF administers 20.6±7.93 Gy, in contrast to VMAT_FFF, which delivers 19.6±7.41 Gy. This difference is also statistically significant (p=0.003), suggesting that VMAT_ FF results in a greater dose to 50% of the parotid gland compared to VMAT_FFF. However, no substantial differences were found in specific dose-volume metrics, such as V50 (11.3±6.95% in VMAT_FF versus 11.6±7.41% in VMAT_FFF, p=0.326), V40 (24.2±10.3% in VMAT_FF versus 24.2±9.72% in VMAT_FFF, p=0.863), and V30 (36.3±12.7% in VMAT_FF versus 35.3±10.9% in VMAT_FFF, p=0.178).

A similar trend was observed for the left parotid gland, where the mean dose was significantly lower in the FFF plan (24.9±5.27 Gy) compared to the FF plan (25±5.19 Gy, p=0.003), For the Left Parotid Gland, the average dose administered by VMAT_FF is 24.9±5.16 Gy, whereas VMAT_FFF delivers an average dose of 24.5±5.27 Gy, with a statistically significant difference observed (p=0.005). Regarding D50, which represents the dose received by 50% of the parotid volume, VMAT_FF provides a dose of 20.7±7.94 Gy, in contrast to VMAT_FFF, which delivers 19.8±7.74 Gy, also showing a significant difference (p=0.037). This suggests that VMAT_FF administers a higher dose to 50% of the left parotid gland compared to VMAT_FFF. While the dose-volume parameters V50, V40, and V30 did not show significant differences.

A comparison of the maximum dose delivered to the eyes revealed no significant difference for the right eye, with values of 3.45±3.37 Gy in the VMAT_FF plan and 3.33±3.66 Gy in the VMAT_FFF plan (p=0.190). However, a significant reduction in the maximum dose to the left eye was observed in the VMAT_FFF plan, which recorded a dose of 5.64±10.5 Gy, in contrast to the VMAT_FF plan"s dose of 5.99±10.7 Gy (p=0.005).

The comparison of NTCP values between VMAT_FF and VMAT_FFF plans indicated that there were no significant differences for most organs at risk (OARs). In particular, the NTCP values for the spinal cord, brainstem and optic chiasm, eye related to myelitis necrosis, necrosis/infarction and blindness were both recorded as 0.00±0.00 for the VMAT_FF and VMAT_FFF plans. Furthermore, although the NTCP values for joint dysfunction in the mandible were slightly higher in the VMAT_ FFF plans (0.686±0.369) compared to the VMAT_FF plans (0.656±0.376), this difference did not reach statistical significance (p=0.072). Concerning the parotid glands, The NTCP values for Xerostomia, derived from the Poisson-LQ model, exhibited minimal differences between VMAT_FF and VMAT_FFF. For the right parotid, the values recorded were 17.8±8.17 and 17.9±8.29, while for the left parotid; the values were 20.3±11.2 and 20.7±11.4. These findings indicate that there is no significant difference, implying that the NTCP values remain consistent across both treatment modalities (p>0.05).

Discussion

Fig. 1. Axial slice showing isodose distribution for VMAT_FF and VMAT_FFF, with isodose lines up to 35% of the maximum dose and color wash depicting overall dose distribution.
VMAT: Volumetric modulated arc therapy; FF: Flattening filter; FFF: Flattening filter free.

Fig. 2. Dose volume histogram (DVH) curves comparing VMAT_FF and VMAT_FFF, showing dose distribution to planning target volumes and organs at risks.
VMAT: Volumetric modulated arc therapy; FF: Flattening filter; FFF: Flattening filter free.

In our study, dosimetric analysis of PTV-60Gy, the maximum and mean doses delivered to the target volume were observed to be comparable in both VMAT_ FF and VMAT_FFF techniques. Furthermore, the conformity index (CI) in the FFF mode demonstrated a significant similarity to that of the FF mode, indicating that the dose distribution conformity between the two methods is alike. These results are consistent with earlier research that has examined the dosimetric properties of flattening filter-free (FFF) linear accelerators. For instance, Zwahlen et al.[] reported that FFF beams, while offering benefits such as reduced treatment duration and enhanced dose delivery efficiency did not markedly change the dose distributions within the target volume when compared to FF beams, especially regarding the mean and maximum doses to the target.

Dosimetric indices of dose distribution, the uniformity index (UI), homogeneity index (HI), gradient index (GI), and undefined dosimetric index (UDI) in our study revealed no significant differences between the two modalities. This suggests that both VMAT_ FF and VMAT_FFF exhibit similar characteristics in dose distribution regarding uniformity and homogeneity. These results align with the findings of earlier studies conducted by Kim et al.[] and Hrbacek et al.,[] which indicated that FFF techniques deliver comparable dose uniformity and target coverage, while also enhancing treatment efficiency through shorter beam delivery times.

A minor variation in the mean dose for the PTV- 54Gy was noted between the two techniques, with VMAT_FF administering a slightly lower dose (54.2±0.217 Gy) compared to VMAT_FFF (54.3±0.217 Gy), which was statistically significant (p=0.011). Although this difference is minimal, it may indicate the influence of the lower energy and the more rapid off-axis dose reduction associated with the FFF mode, as previously highlighted in the research conducted by Sarma et al.[] and Low et al.[]. These studies proposed that FFF beams might encounter difficulties in achieving the same depth of dose distribution in larger target volumes due to the accelerated dose fall-off, thereby complicating the delivery of uniform dose coverage across the entire target.

Monitor units (MU) employed in FFF VMAT planning were observed to be significantly higher than those used in FF VMAT planning (583±52.1 compared to 530±69.9, p=0.001). This observation aligns with previous studies, such as that conducted by Zwahlen et al.,[] which suggested that FFF beam delivery requires a greater number of MUs due to the absence of a flattening filter. However, despite the elevated MU associated with FFF plans, no significant difference in treatment duration was observed between the two planning methods. This can be attributed to the administration of a consistent single dose of 2Gy in both FFF and FF modalities, which alleviated any potential impact of MU discrepancies on treatment time. These findings indicate that while FFF plans necessitate a higher quantity of MUs, they do not compromise therapeutic effectiveness.

Figure 3 illustrates the correlation between the Equivalent Dose in 2 Gy/Fraction (Gy) and the LQ Scaled DVH, which serves as an indicator of radiation dose distribution, for multiple types of Normal Tissue Complication Probabilities (NTCPs) in a single patient. The graph presents the NTCP curves corresponding to various organs and tissues, such as the Mandible, Spinal Cord, Brain Stem, Parotid Gland, among others. A significant insight derived from this graph is the divergence observed in the NTCP curves across various organs and tissues, which reflects their distinct sensitivities to radiation exposure. The NTCP curves for the Spinal Cord and Brain Stem exhibit steeper inclines, implying that these tissues are more susceptible to increases in radiation dosage. In contrast, the NTCP curves for the Mandible and Parotid Gland display shallower slopes, suggesting that these tissues possess a greater resistance to radiation- induced damage. This variation in sensitivity is a crucial factor in treatment planning, as healthcare professionals must strive to deliver an effective therapeutic dose to the tumor while concurrently minimizing the potential for unacceptable toxicities to adjacent healthy tissues.

Fig. 3: Linear-quadratic (LQ) scale DVH showing biologically scaled dose-volume histograms for OAR structures used in NTCP analysis.
DVH: Dose volume histogram; OAR: Organ at risks; NTCP: Normal tissue complication probability.

The graph further illustrates that specific normal tissues, including the Parotid Gland and Xerostomia, exhibit two separate NTCP curves, which are distinguished by the labels VMAT_FF and VMAT_ FFF. This indicates that the selection of the radiation beam whether a VMAT_FF or a VMAT_FFF beam can influence the likelihood of normal tissue complications in these organs for the patient. The variation between the NTCP curves for VMAT_FF and VMAT_FFF suggests that the type of beam may significantly affect the risk of toxicities, with the VMAT_FFF beam potentially presenting a reduced probability of complications for certain organs when compared to the VMAT_FF beam at the same equivalent dose level. These results underscore the critical role of beam selection in radiation treatment planning and emphasize the necessity of integrating this understanding into the optimization process to enhance patient outcomes and quality of life.

Organ-specific Finding
The mean dose and D50 for the left parotid gland were decreased by 1.61% and 4.35%, respectively, in the VMAT_FFF technique when compared to VMAT_FF. However, the NTCP Poisson-LQ (Xerostomia) exhibited a slight increase of 1.93%, which was not statistically significant. In a similar manner, the right parotid gland experienced reductions of 2.03% in mean dose and 4.85% in D50, accompanied by a non-significant increase of 0.56% in NTCP. In alignment with previous research, doses to the parotid glands that exceed 25-30 Gy significantly elevate the risk of xerostomia. [-]. Our results further support the notion that minimizing parotid doses can lead to a decrease in xerostomia incidence, underscoring the critical need for dose optimization to mitigate side effects and enhance treatment outcomes.

Mandibular dysfunction, which encompasses conditions such as trismus and osteoradionecrosis, exhibits a strong association with the maximum radiation dose absorbed by the mandible. Previous studies have indicated that doses surpassing 60-65 Gy significantly elevate the likelihood of these complications.[,]. In our evaluation found the maximum dose delivered to the mandible in the FF plan exhibited a negligible reduction of 0.16% when compared to the FFF plan, accompanied by a 4.37% decrease in the NTCP for joint dysfunction. These results indicate that the risk profiles for mandibular dysfunction remain consistent between the two techniques; however, the marginally elevated NTCP observed in FFF plans necessitates additional examination.

Kirkpatrick et al.[] conducted a study on the radiation tolerance of the spinal cord and determined that the likelihood of developing myelitis necrosis significantly increases when radiation doses surpass 45-50 Gy using conventional fractionation (2 Gy per fraction). This dosage threshold indicates the onset of irreversible damage to the spinal cord, which can result in complications such as radiation myelopathy. Their research underscored the necessity of adhering to stringent dose limits to reduce neurological risks. In our investigation, both VMAT_FF and VMAT_FFF techniques demonstrated comparable maximum doses to the spinal cord, achieving a dose reduction of 29.33% in relation to the established tolerance limit.

The brainstem, similar to the spinal cord, exhibits a restricted tolerance to radiation owing to its essential neurological roles. Doses surpassing 54 Gy markedly elevate the likelihood of necrosis and vascular injury.[,] In this study, the NTCP values associated with brainstem complications were minimal, with the maximum doses in both FF and FFF plans remaining within permissible thresholds, thereby providing adequate protection to the brainstem without undermining the therapeutic dose. These results underscore the necessity of following dose limitations to reduce radiation-related damage while preserving clinical effectiveness.

The optic chiasm exhibits a significant sensitivity to radiation, with a tolerance dose estimated at around 55 Gy when utilizing conventional fractionation (2 Gy per fraction) to reduce the likelihood of complications such as blindness. Clinical research, including the work of Emami etal.,[] indicates a TD5/5 of 50 Gy for the optic chiasm. These parameters have become essential components of contemporary radiotherapy protocols aimed at preventing radiation-induced optic neuropathy. Surpassing these limits can lead to irreversible damage to the optic pathways, highlighting the importance of adhering to dose limitations during the treatment planning process. In our study, the maximum dose (Gy) in the VMAT_FFF plan demonstrated a 2.78% decrease compared to the VMAT_FF plan; however, this difference lacks statistical significance. The NTCP Poisson-LQ (Blindness) revealed no difference, with both values recorded at 0.

Radiation exposure has the potential to cause ocular complications, including cataracts, retinopathy, and optic neuropathy. Cataracts are generally observed at radiation doses exceeding 2 Gy, while optic neuropathy is associated with doses greater than 45 Gy. To mitigate these risks, radiotherapy protocols are designed to keep the maximum dose to the eyes below 45 Gy. In our investigation, the maximum doses administered to both the right and left eyes remained within these established limits, suggesting a minimal risk of clinically significant visual impairment. This finding is consistent with guidelines that advocate for rigorous dosimetric control to safeguard ocular structures while delivering effective treatment. Research conducted by Mayo et al.[] and Bhandare et al.[] reinforces these dose limitations, underscoring the necessity of keeping doses significantly below 50 Gy to prevent complications. Furthermore, studies referenced asand have identified comparable thresholds for radiation-induced ocular toxicity.[] The percentage difference in maximum doses for both eyes were negligible, with the right eye receiving 3.48% and the left eye 5.84% lower doses in VMAT_FFF compared to VMAT_FF, while the NTCP Poisson-LQ for blindness indicated no difference (0%).

Figure 4 illustrates the comparison of the Gamma Passing Rate (GPR) between VMAT_FF and VMAT_ FFF for a cohort of 20 patients. Both techniques exhibited clinically acceptable GPR values exceeding 97%. However, VMAT_FF consistently demonstrated marginally higher passing rates than VMAT_FFF. This indicates slightly enhanced delivery accuracy with FF beams, although both methods met high standards for quality treatment delivery.

Fig. 4. Gamma passing rate (%) comparison between VMAT_FF and VMAT_FFF.
VMAT: Volumetric modulated arc therapy; FF: Flattening filter; FFF: Flattening filter free.

Limitation of the Study
The retrospective nature of the study, along with its restricted sample size, limits both its statistical power and the ability to generalize the findings. Although the Normal Tissue Complication Probability (NTCP) analysis employing the Poisson Linear Quadratic (LQ) model offers important insights, it may not fully encompass the intricate nature of clinical toxicity.

Conclusion

Ethics Committee Approval: The study was approved by the Indira Gandhi Institute of Medical Science Office Ethics Committee (no: 35/IEC/IGIMS/2024, date: 17/05/2024).

Conflict of Interest: The authors have no conflicts of interest to declare.

Financial Support: The authors declared that this study received no financial support.

Use of AI for Writing Assistance: No AI technologies utilized.

Authorship Contributions: Concept - M.Z., D.P.; Design - D.P., M.Z.; Supervision - M.Z., D.P.; Data collection and/or processing - D.P., M.Z.; Data analysis and/or interpretation - D.P., M.Z.; Literature search - D.P., M.Z.; Writing - D.P., M.Z.; Critical review - M.Z., D.P., D.S.

Peer-review: Externally peer-reviewed.

References