Effect of Jaw Tracking During Volumetric Modulated Arc Therapy for Facial Non-melanoma Skin Cancer

Abstract

Background/Aim

This study aimed to analyze the dosimetric effects of jaw tracking during Volumetric Modulated Arc Therapy (VMAT) planning for facial non-melanoma skin cancer (NMSC).

Patients and Methods

This study included 50 patients with facial NMSC who underwent VMAT planning with or without jaw tracking. The target volume (TV) included the primary skin lesion with a 1-cm margin around the surface and a depth of 4 mm. A total of 55 Gy in 20 fractions was prescribed, and the plans were considered acceptable if the TV was covered by 95-105% of the isodose curve. A dosimetric comparison was performed for the volumes of the low-dose regions, which were defined as <50% of the prescription dose (V10-50%). Target coverage was evaluated using the homogeneity index (HI) and conformity index (CI).

Results

The patients’ mean TV was 5.137 cc (range=1.03-15.89 cc). Jaw tracking resulted in mean volume reduction rates of 3.9%, 6.6% 10.6% and 13.8% for V40%, V30%, V20%, and V10%, respectively (all p<0.001). The volume change in V50% between the two groups was 2.7% (p=0.006). No significant differences were observed in HI (p=0.449) or CI (p=0.127).

Conclusion

The application of jaw tracking during VMAT for facial NMSC is associated with a significant reduction in the volume of low dose delivered in the radiation field (V10-50%), while maintaining target coverage. Future analyses should assess whether this volume difference affects treatment-related cosmetic outcomes.

Radiation therapy (RT) is preferred for elderly or frail patients who are not candidates for primary surgery for non-melanoma skin cancer (NMSC) (1). In addition, RT should be offered for lesions of the central face, lips, eyelids, and ears if surgery leads to inferior cosmetic or functional outcomes (2,3). Intensity-modulated RT (IMRT) with hypofractionation (<20 fractions) has allowed tighter margins around target volumes and higher radiation dose delivery, possibly leading to improved accuracy and treatment tolerance (3). However, a higher daily dose is associated with worse cosmetic results.

Jaw tracking is a radiation transfer method that minimizes leakage and transmission from the leaves while keeping the jaw closest to the aperture of a multileaf collimator (MLC). If jaw tracking is used, transmission can be reduced to 0.1% or less (4). Although RT is a tissue-preserving modality that may offer better cosmetic outcomes, the higher extent of low-dose radiation field can have detrimental effects on the skin, such as erythema and moist desquamation, especially for facial NMSC (5). Therefore, it is necessary to reduce treatment-related skin complications by reducing the low dose range of radiation.

This study aimed to evaluate the dosimetric effects of the jaw tracking technique by analyzing the low-dose extent outside the treatment area with or without jaw tracking during volumetric modulated arc therapy (VMAT) for facial NMSC.

Patients and Methods

Treatment planning of facial skin cancer patients. Data from the scans of 50 patients with facial NMSC were transferred to the Eclipse™ Treatment Planning System (Version 16.0; Varian Medical System, Palo Alto, CA, USA). All patients underwent VMAT planning with or without tracking, comprising 100 plans (Figure 1). The target volume (TV) included the primary skin lesion with a 1 cm margin around the surface and a depth of 4 mm. A bolus was applied to the tumor surface. A total of 55 Gy in 20 fractions was prescribed, and the plans were considered acceptable if the TV was covered by 95%-105% of the isodose curve. The plans using 6MV photon beams on the VitalBeam equipped with the Millennium 120 MLC (Varian Medical System) were calculated. To minimize the variables in the process of optimizing the plan for the same patient, the same dose constraint was applied to the virtual ring around the target without applying the OAR (organ-at-risk) dose constraint when applying Jaw tracking. This study was approved by the Jeju National University Hospital Institutional Review Board (No. 2023-08-022) and was performed in accordance with the Declaration of Helsinki.

Figure 1. The target volume (TV) included the primary skin lesion with a 1-cm margin around the surface (A). The plans with (B) or without Jaw tracking (C) were considered acceptable if the TV was covered by 95-105% of the isodose curve.

Dosimetric analyses. TV coverage was evaluated via the conformity number (CN) and homogeneity index (HI) using the definitions below:

,

where TV_95% is the target volume covered by at least 95% of the prescribed dose, TV is the target volume, and V_95% is the volume of the 95% isodose. D_2%, D_50%, and D_98% are the doses received at 2%, 50%, and 98% of the target volume, respectively.

To evaluate the effect of Jaw tracking on radiation leakage, low-dose regions were defined as dose distributions under 50% of the prescribed dose. Dosimetric analyses were performed using volumes in the low-dose region (V10%, V20%, V30%, V40%, and V50%). The effects of radiation delivery on jaw tracking were evaluated using quality assurance (DQA). One hundred treatment plans were delivered using portal dosimetry (Varian version 16.0), and gamma analysis was performed at a 3 mm/3% baseline with a threshold of 10%. Differences between the low-dose regions with and without the jaw tracking were compared using a paired t-test. Spearman’s correlation test was used to determine the association between dosimetric parameters. All statistical analyses were performed using SPSS version 20.0 (SPSS Corp., Armonk, NY, USA). Statistical significance was set at p≤0.05.

Results

Patients. Detailed clinical information regarding the patient and tumor characteristics is presented in Table I. The median age at the time of planning was 79 years (range=49-98 years). Thirty-four patients were female, and 16 were male. The most common tumor locations on the face were the cheek (n=16) and nose (n=15), followed by the periorbital region (n=4). Histological examination revealed Basal cell carcinoma (BCC) in 28 patients, Squamous cell carcinoma (SCC) in 14, and SCC in situ (Bowen’s disease) in 8. Thirty-five, six, and one of the patients were classified as having stages I, II, and III disease, respectively. The patients’ mean TV was 5.137 cc (range=1.03-15.89 cc).

Table I. Patient and tumor characteristics.

Dosimetric comparisons between groups with and without jaw tracking. Figure 2 shows the volumes for the dose distributions of 10%, 20%, 30%, and 40%. Jaw tracking resulted in mean volume reduction rates of 3.9%, 6.6%, 10.6% and 13.8% for V40 %, V30%, V20%, and V10%, respectively (all p<0.001). The volume change in V50% between the two groups was 2.7% (p=0.006). There were no significant differences above 50% dose distribution, except for V90% (Table II). The average Monitor Unit (MU) value was 683.68 and 669.44 for plans with and without jaw tracking, respectively.

Figure 2. Volume for low dose distribution (V10%-V40%) of groups with or without Jaw tracking. All parameters were significantly different between groups (p<0.001).

Table II. The mean volume (cc) of the dose distribution above 50% between with and without jaw tracking.

No significant differences were observed in the HI (p=0.449) or CI (p=0.127), as shown in Figure 3. The mean (standard deviation: SD) of HI and CI on the target were 0.09 (±0.08) and 0.54 (±0.07) with jaw tracking, and 0.09 (±0.05) and 0.54 (±0.07) without Jaw tracking, respectively. When performing DQA, all plans met the gamma evaluation requirement of >95%, and there were no significant differences between the two techniques.

Figure 3. Comparison by homogeneity index (HI) and conformity index (CI) of all patients with or without jaw tracking to evaluate target coverage.

Discussion

Hypofractionated radiation therapy for NMSC has the advantage of improving the patients’ quality of life and enabling treatment without surgery (6). A meta-analysis of hypofractionated RT for NMSC involving 9,729 patients reported median local recurrence rates of 2% at 1 year and 14% at 5 years (7). Especially in frail elderly patients (aged ≥80 years) who are not amenable to undergoing surgery because of comorbidities, hypofractionated RT with 20 fractions for NMSC showed good clinical efficacy without discontinuing the planned course of RT (8). However, there would be side effects of radiation including radiation dermatitis on the skin; thus, it is important to control the low-dose area around the target to increase the effectiveness of radiation therapy.

The higher the extent of the low-dose radiation field, the worse the skin erythema. The jaw tracking technique was developed to reduce the leakage and transmission dose of the MLC with existing linear accelerators (9). The position of the MLC within the radiation field has a significant impact on the leakage and transmission and is also related to the surface dose. The leakage and transmission through the MLC leaves in the jaw tracking technique are minimized because the jaw’s position is as close as possible to the MLC aperture (10,11). Comparison of measurements on the surface dose of 6MV photon beam among three fields (jaw only vs. MLC only vs. MLC+jaw) using Varian’s 120 MLC resulted in a dose reduction of approximately 2-3% at the field created using both MLC and jaw (12). Jaw tracking contributes to the dose below 10% by reducing leaf transmission (13). Our results showed a significant volume reduction of about 13.8% at doses below 10%, and this effect was also evaluated to be present in 20-40% (3.9-10.6%) of the dose distribution. In addition, compared with the case where jaw tracking is not applied, it has the efficiency of radiation delivery without a difference in the MU. Several studies have shown that jaw tracking can maintain target coverage and reduce the OAR dose. In the comparison of the dose to the OARs, a maximum difference of 2% was observed between the two techniques in the case of including OAR within the treatment field (14-16). There was no difference between the CI and HI for the target, and the average MU was 14.24 in this study. Although we did not apply any contours, we were able to improve the conformity of the target and control the dose distribution, except in low-dose regions. Although the mean volume increased by 1.9% in the V90% with jaw tracking, as shown in Table II, it can be adjusted using ring structures around the target volume and constraints for the optimization of treatment planning (17).

This study used two comparative planning designs and included homogeneity at the treatment site for facial NMSC. We verified the gamma evaluation by irradiating rats with radiation according to the application of jaw tracking. There was no difference in terms of DQA results between the two techniques. In addition, our study of many planned cases showed that the use of jaw tracking for NMSC reduces the low-dose volume, which is not arbitrarily adjusted in the radiation treatment planning process, while maintaining the target and other factors related to radiation delivery, including MU and DQA. However, in vivo dosimetry of the skin in the low-dose area and clinical evaluation of the dose-skin response are required to confirm the clinical effect of the jaw tracking technique. Thus, it may be possible to obtain information on which part of the low-dose region has a strong clinical effect.

Conclusion

The application of jaw tracking during IMRT for facial NMSC is associated with a significant reduction in the volume of the low dose delivered in the radiation field (V10-30%) while maintaining the target coverage. Future clinical analyses should assess whether this volume difference affects treatment-related cosmetic outcomes.

Funding

This work was supported by a research grant from the Jeju National University Hospital in 2023 (grant no. 202300150001).

Conflicts of Interest

The Authors declare that they have no conflicts of interest in relation to this study.

Authors’ Contributions

Study conception and design: So Hyun Park, Jinhyun Choi; Acquisition of data: So Hyun Park, Hyunsoo Ko; Analysis of data: So Hyun Park, Jinhyun Choi; Article draft: So Hyun Park, Hyunsoo Ko, Jinhyun Choi; Article revision: So Hyun Park, Jinhyun Choi. All Authors read and approved the final article.