Interfractional change of high-risk CTV D90 during image-guided brachytherapy for uterine cervical cancer

摘要

Between October 2008 and October 2010, 61 patients with previously untreated uterine cervical cancer were treated with radiotherapy with radical intent at our institution. In this study, we analyzed the patients who were treated with high-dose-rate intracavity brachytherapy (HDR-ICBT) using a combination of tandem and ovoid applicators. Nine of the patients were ruled out, as one was treated with a vaginal cylinder and eight were treated with interstitial brachytherapy. Therefore, 52 patients were eligible for this analysis (age range, 33–84 years; median, 63 years), and their performance status was 0–2. Patient characteristics are shown in Table 1. All patients underwent MRI of the pelvis for pretreatment evaluation, and cervical tumor dimensions were measured based on T2-weighted images. All patients were treated with a combination of EBRT and HDR-ICBT. In principle, chemotherapy was concurrently combined in patients with tumors 4 cm in FIGO Stage I–II, in FIGO Stage III–IVA, or with pelvic lymph-node metastases. The exclusion criteria for chemotherapy were age over 75 years or severe concomitant diseases. Of the 52 patients, 23 (44%) were treated with concurrent chemoradiotherapy: 21 patients with weekly cisplatin at 40 mg/m2 (2–5 times), and two patients with weekly cisplatin at 30 mg/m2 and paclitaxel at 50 mg/m2 (6 times). All contouring (HR-CTV, bladder and rectum) was reviewed by two radiation oncologists (Y.O. and T.O.). Cumulative dose–volume histograms (DVHs) were calculated for the HR-CTV and the OARs. The minimum doses delivered to 2 cm3 (D2cc) of the most irradiated volumes of the bladder and rectum were derived from the DVH according to the GEC-ESTRO recommendation [3]. From DVH analysis, all parameters of the HR-CTV volume, HR-CTV D90, D2cc of bladder, and D2cc of rectum were obtained for four fractions of HDR-ICBT for each patient. We analyzed these changes in three subgroups based on tumor size: small tumor, ≤4 cm in maximum diameter in MRI images at pretreatment; medium tumor, 4–6 cm; large tumor, ≥6 cm. Similarly, we analyzed two subgroups based on the HR-CTV volume at first ICBT: small group, 35 cm3 in HR-CTV volume; large group, ≥ 35 cm3. The reason for setting the cut-off value at 35 cm3 was that the HR-CTV volume is estimated to be ～ 35 cm3 in the case of 4-cm tumors of maximum diameter, assuming the cervical tumor to be a spherical object. Retrospectively, these patients were divided into two groups. Of the 52 patients in the study, 28 were prescribed 6 Gy per fraction at point A in each ICBT for four fractions with the standard loading pattern of our practice (‘standard dose group'). In this group, there was no dose modification for OARs, as the intestinal dose achieved our dose constraints based on visual inspection of the isodose lines to the intestine. An unpaired Mann–Whitney U test was used to compare the dose between the ‘standard dose' and ‘adaptive dose' groups. All P values reported were two-sided; P 0.05 was considered statistically significant. Figure 2 shows the changes over time of the point A dose and the HR-CTV D90 from the first to fourth HDR-ICBT. In the standard dose group, the point A dose was always 6 Gy. On the other hand, the HR-CTV D90 increased through all four courses of ICBT, especially from the first to third ICBT: 6.1 Gy on average (SD = 1.3) in the first ICBT, 6.6 Gy (SD = 1.0) in the second, 7.0 Gy (SD = 1.0) in the third, and 7.1 Gy (SD = 1.0) in the fourth. The total HR-CTV D90 of EBRT and ICBT was 65.0 Gy_EQD2 on average (SD = 7.3). Figure 3 shows the interfractional changes over time of the HR-CTV D90 and HR-CTV volume, presented in each pretreatment tumor size group (small, medium, and large). In the medium (4–6 cm) and large (≥6 cm) tumor groups, the values of the HR-CTV D90 at first ICBT in the standard dose group were significantly lower than those in the adaptive dose group (P = 0.035 and P = 0.017, respectively). In each size group the total HR-CTV D90 of EBRT and ICBT tended to be higher in the adaptive dose group than in the standard dose group, although the differences were not statistically significant. Regarding the volume of the HR-CTV, the small and large tumor groups showed relatively lower volumes in the adaptive dose group compared with the standard dose group, although there was no statistical significance indicated. Figure 4 shows the interfractional changes of the HR-CTV D90 and volume, presented in each HR-CTV volume at first ICBT (small and large). In the large tumor group (≥35 cm3), the HR-CTV D90 values at first, second and third ICBT in the adaptive dose group were significantly higher than those in the standard dose group (P = 0.008, 0.004 and 0.037, respectively). The total HR-CTV D90 of EBRT and ICBT in the large tumor groups was 61.5 Gy_EQD2 on average (SD = 5.6) in the standard dose group