Patient data and setup

This retrospective study included a total of 63 women with cervical or endometrial cancer who underwent adjuvant radiotherapy at the Third Affiliated Hospital of Sun Yat-sen University between August 2020 and November 2022. The study has been approved by Medical Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University (Prot. Number II2023-149-01). The patients’ ages ranged from 27 to 74 years, with a median age of 52 years. All patients had a confirmed pathological diagnosis and were eligible for radiotherapy without any contraindications. For immobilization, all patients were positioned in the prone position using the PIDBB system, with their hands raised upward to grip the handle and their chest and abdomen supported by a flat frame. The PIDBB system used in this study (Klarity Inc, Guangzhou, China) consisted of a carbon fiber floor, engineering foam support pad, negative pressure vacuum bag, and thermoplastic film (Fig. 1a). The effectiveness and reproducibility of the PIDBB system were previously validated in our institution [14, 15]. Before each fraction of radiotherapy, bladder urine volume measurements were performed for all pelvic patients to ensure accurate positioning. Throughout the simulation and treatment process, all patients maintained the same position. CT images of each patient were acquired using a large aperture CT simulation device (SOMATOM Definition, Siemens Healthiness, Germany, Munich). The scan range extended from the 11th thoracic vertebra to 5 cm below the ischial tuberosity, with a slice thickness of 3 mm. The acquired CT images were then transferred to the treatment planning system (Monaco V6.0, Elekta AB, Stockholm, Sweden) via the radiotherapy network for treatment planning (Fig. 1b and c).

Fig. 1

Typical CT image with PIDBB system. (a). The newly developed PIDBB system consists of several components, including a carbon fiber floor (black part), an engineering foam support pad (green part), a negative pressure vacuum bag (blue part), and a thermoplastic film (white part). (b) Transverse plane CT images of a patient immobilized with PIDBB system. The blue contours represent the carbon fiber floor, the yellow contours represent the engineering foam support pad, the green contours represent the negative pressure vacuum bag, and the purple contours represent the thermoplastic film. (c) Sagittal plane CT image of a patient immobilized with PIDBB system

Regions of interest

The gross target volume (GTV) for gynecologic cancer was determined based on gynecological examination and imaging diagnosis. The clinical target volume (CTV) was determined by considering the direct spread and lymph node metastasis pathways specific to cervical or endometrial cancer. The planning target volume (PTV) was obtained by expanding the CTV outward by 5 mm to account for setup errors. The OARs included the small intestine, colon, rectum, bladder, femoral heads, and skin. The radiation therapy structures were delineated on CT images for all patients by an experienced radiation oncologist, following the guidelines outlined in the National Health Commission of the People’s Republic of China Standard Practice for Diagnosis and Treatment of Cervical Cancer (2018 edition), National Health Commission of the People’s Republic of China Endometrial Carcinoma Diagnosis and Treatment Specification (2018 edition), and International Commission on Radiation Units and Measurements (ICRU) Report 83 [16]. To evaluate the variation in surface dose caused by the immobilization devices, the skin contour was delineated with a thickness of 3 mm below the skin surface for each patient. In this study, two sets of external body contours were created for each patient: one set included only the patient’s body without the PIDBB, while the other set included the patient’s external body contours along with the PIDBB.

Treatment planning and dose calculation

The purpose of this study was to evaluate the impact of using a patient-specific bolus (PIDBB) on the dose distribution in multi-beam IMRT plans. The plans were designed using 6 MV X-ray beams from a medical linear accelerator (Synergy, Elekta AB, Stockholm, Sweden). Seven evenly distributed coplanar beams (150°, 100°, 50°, 0°, 310°, 260°, 210°) were used for each plan. The optimization of the plans was performed using the Monte Carlo (MC) algorithm combined with dynamic multi-leaf collimator (dMLC) technology. Each field control number was set to 20, and the calculated grid size was 3 mm × 3 mm × 3 mm. The uncertainty of each control point of the MC algorithm was set at 3%. The maximum and minimum doses were planned according to the recommendations of ICRU Report 83 [16], with dose constraints based on the Quantitative Analyses of Normal Tissue Effects in the Clinic (QUANTEC) guidelines [17].

Two IMRT plans were generated for each patient, one with the PIDBB (Plan+) and one without (Plan−). The prescription dose to the PTV was 50 Gy, delivered in 25 fractions. To evaluate the skin dose, interference from target coverage was eliminated by renormalizing the dose distribution in Plan− and Plan+ to achieve a clinically desired prescribed dose of V50 Gy = 95%. These renormalized plans were recorded as Plan− n and Plan+ n, respectively.

Statistical analysis

Dose-volume histograms (DVHs) are commonly used to assess the dose coverage of PTVs and OARs. In this study, several parameters were used to evaluate the dose distribution within the PTVs. These parameters included the mean dose (Dmean), the homogeneity index (HI), and the conformity index (CI). The HI and CI were calculated using the following formulas [18, 19]:

$$\:CI=\frac_}_}*\frac_}_}$$

(2)

In formula (1), D2%, D50%, and D98% represent the doses received by 2%, 50%, and 98% of the volume of ROI, respectively. According to the ICRU Report 83, D2% represents the near maximum dose, while D98% represents the near-minimum dose within the ROI. A smaller HI value, closer to 0, indicates a more uniform dose distribution within the target volume.

In formula (2), VT represents the volume of the target, VRX represents the volume of the target covered by the reference isodose curve (which was set at 100% of the prescribed dose in this study), and VRI represents the total volume within the reference isodose curve. The CI ranges from 0 to 1, with higher CI values indicating better dose conformity to the target volume.

To assess the impact of the immobilization device on the overall treatment plan, the TPS was used to calculate the perturbation effect. This was done by subtracting Plan- from Plan+, and the average of the parameter differences between the two plans was represented as \(\:\stackrel\) by the following formula (3).

$$\:\stackrel=_^\left[_-_\right]/63$$

(3)

IBM SPSS Statistics (version 22.0, IBM Corp., Armonk, NY, USA) was used to analyze all data. Paired-samples T tests were employed to determine the significance of the observed differences between Plan+ and Plan− as well as between Plan+ n and Plan− n. AP-value of less than 0.05 was considered statistically significant for these differences.

Dose verification of the anthropomorphic phantom

An Alderson Radiation Therapy Anthropomorphic Phantom (ARTAP) (RSD Inc., USA) and gafchromic EBT3 film (ASHLAND Inc., USA) were used to verify the dose distribution of Plan− and Plan+. The ART phantom used in this study is composed of a tissue-equivalent material that closely resembles the scattering and absorption properties of human tissue. The phantom is horizontally divided into slices that are 2.5 cm thick, replicating the anatomy of the abdomen and pelvis (Fig. 2a). EBT3 films, obtained from the same batch (NO. 09131801) with a calibration dose range from 0 Gy to 4 Gy, were used for dose measurements. These films were cut into appropriate sizes and placed between the transverse slices of the phantom. To simulate the actual patient treatment process, the phantom was immobilized using the PIDBB and scanned using a large aperture CT, similar as a real patient. The actual radiotherapy plan was executed using a linear accelerator, and the dose value of the hypogastrium skin (point A, Fig. 2b) was obtained using the EBT3 film in the ART phantom. This study used 90 × 75 × 0.34 mm flims for dosimetric measurements. 189 flims were used for skin dose measurement in each plan. After taking three measurements for each plan, the averages of these measurements were calculated. The calculated mean dose was multiplied by 25 to arrive at the total dose. The irradiated films were then scanned and imported into FilmQA Pro 2016 software (ASHLAND Inc., USA). In the Monaco TPS, a verification plan was created for Plan− to obtain the dose value at point A, using the CT images of the ART phantom as a verification phantom. Similarly, dose values mapped from Plan+ to point A were obtained. The differences between the dose measured on the irradiated EBT3 film and the TPS calculations from Plan + and Plan- were then determined.

Fig. 2

Typical image of anthropomorphic phantom. (a) An anthropomorphic phantom was utilized for film dosimetry. This phantom was horizontally divided into slices that were 2.5 cm thick, replicating the anatomy of the abdomen and pelvis in actual patients. The phantom was fixed using immobilization devices to simulate the complete real treatment process, with the exception of removing the thermoplastic film for photographic convenience. (b) EBT3 films were placed between the transverse slices of the phantom, creating a sandwich-like arrangement. These films were used to measure the dose distribution within the phantom

Dose verification of patient-specific quality assurance (QA)

Patient-specific verification QA plays a crucial role in identifying any discrepancies between the calculated and delivered radiation doses. The Amerian association of physicists in medicine (AAPM) TG-218 report [20] provides universal action limits for such QA, stating that the γ passing rate should be equal to or greater than 90%, with a criterion of 3%/2 mm and a 10% dose threshold. For this study, it is necessary to perform verification QA for both Plan+ and Plan− for a total of 63 patients. This is done to assess the accuracy of radiation delivery to these patients and ensure that the treatment plans are being executed as intended.