Discussion
Brachytherapy is an essential part of cervical cancer treatment. It is able to provide the maximal dose to the target area while minimizing the dose to OARs. Previous studies demonstrated that ICBT is a safe, effective modality for cervical cancer28. However, tumors with large residual disease after EBRT present a therapeutic challenge. Attempts have been made to improve tumor dose coverage by ICBT. Optimization of pear shaped isodose configuration is an easy and simple method but it is limited because of the limited number of source position. Other dosimetric studies have investigated the possibility replacing ICBT with IMRT or stereotactic body radiotherapy (SBRT)29-31. One must also consider the uncertainty of interfraction and intrafraction motion of the target due to bladder and rectal filling30-32. Hybrid IC/ISBT is a relatively new technique using intra-cervical interstitial brachytherapy to improve the target volume coverage8. Multi-institutional clinical studies, retroEMBRACE and EMBRACE-I, have investigated a large number of patients and confirmed favorable clinical outcome and acceptable toxicity 20, 33. EMBRACE-II was launched as a prospective study based on the outcome results of retroEMBRACE and EMBRACE-I with IC/ISBT34.
In 2008, Duan et al 23and Assenholt et al 35reported that a combination of ICBT with IMRT boost maintained the dose distribution and characteristics of ICBT while adequate tumor dose coverage was achieved by supplementary IMRT (IC/IMRT). Comparative dosimetric studies between conventional ICBT, optimized 3-D ICBT and IC/IMRT confirmed that IC/IMRT has better target volume coverage for large tumors while maintaining low dose to OARs35-37. One unique feature of IC/IMRT is this treatment is carried out with the applicator in situ to provide spatial registration and immobilization of the gynecologic organs. After treatment of the ICBT, the supplementary IMRT plan is executed immediately at the same position to ensure the accuracy of target irradiation during the entire treatment process.
Our study included only patients with unfavorable large residual tumor at the time of ICBT. Therefore, Mean HR-CTV volume was a much larger volume (65.8 ± 23.6cc) than other studies. The mean HR-CTV volume was 34 ± 17cc in Kirisits et al study38, 38 ± 20cc in Tanderup et al study39, and 55 ± 38cc in Jurgenliemk-Schulz et al study40.Ken Yoshida et al 41demonstrated that classical conventional ICBT is suitable for the treatment of most HR-CTV size of 36cc or smaller tumors. For bulky tumors patients who poorly responding to EBRT may increase the need for more sophisticated brachytherapy, such as comprehensive interstitial techniques, due to unfavourable geometry at the time of brachytherapy implant. Therefore, it would seem reasonable that IC/IMRT is an alternative to the IC/ISBT or ISBT modality to deliver the boost of dose to the bulky tumors.
The DVH parameters of cervical cancer radiotherapy are related to the local control rate and side effects. Dimopoulos et al42 retrospectively analyzed 141 cervical cancer patients with 51 months median follow-up and demonstrated that the HR-CTV D90 > 87Gy resulted in a LC rate of 96% compared to 80% for HR-CTV D90 < 87Gy. Pötter et al 8 reported that the average D90 dose of HR-CTV in 156 patients treated with HDR-ICBT was 93 Gy, resulting in a 3-year LC rate of 95%. Lindegaard et al43 reported that patients treated by HDR-ICBT with the average HR-CTV D90 doses of 91 Gy had an actuarial 3-year pelvic control rate of 85%. The retroEMBRACE study44 showed that with the systemic usage of IC/IS the D90 of HR-CTV increased 9Gy from 83 ± 14Gy to 92 ± 13Gy and 3-year local control rate in patients having a HR-CTV ≥ 30cm3 was 10% higher in IC/IS group.
In our study, the average D90 doses of HR-CTV and IR-CTV in 76 cases were 88.7 Gy and 78.1 Gy, respectively. The V100 of HR-CTV was more than 90%. Estimated Local recurrence free survival rate at 5 years was 87.6% for all investigated patients, respectively. Estimated metastasis recurrence free survival rate at 5 years was 82.4%. The estimated overall actuarial cancer specific survival at 5 years was 76.3%, the disease free survival at 5 years was 70.9%, respectively.
The side effects of image-guided brachytherapy have been relatively low. The EMBRACE studies showed that a D2cc ≥ 75 Gy was associated with 12.5% risk of fistula and 2.7% with low dose at 3 years. A D2cc < 65 Gy was associated with a two times lower risk of proctitis than ≥ 65 Gy in 960 patients45.  Grade 3 to 4 urinary morbidity was 5.3% in 1176 patients46.  French STIC prospective study 19showed that 3-D imaging based plan reduced Grade 3 to 4 toxicities than 2-D plan, 2.6% and 22.7% respectively in 117 patients. Our study showed that the crude grade 3 late toxicities were 3.9% gastrointestinal and 5.2% urinary system. Neither crude grade 4 acute nor late toxicities were found in gastrointestinal and urinary systems. Actuarial rate for G3 + G4 morbidity was 2.6%/3.9% for the GI, 3.9%/5.2% for the GU at 3/5 years, respectively. Our mean D2cc for OARs was comparable with other studies (bladder 71.8, rectum 64.6, sigmoid colon 63.9 and intestine 56.7 Gy).
Several limitations in our study should be acknowledged. This is a retrospective and single institutional study. Also, due to time limitation, an IMRT-QA check was not done before the IMRT treatment was delivered.
Based on our experience, we would like to emphasize the following points. The IC/IMRT technique can be logistically challenging if the procedure is not orchestrated in proper order. First, the time between ICBT and IMRT should be as short as possible to maintain applicator stability and patient comfort. Second, complementary IMRT was designed to compensate the dose of ICBT due to the lack of dose coverage from ICBT alone, thus the ICBT always contributed the majority of the dosage in this technique, which is important to reduce the dose for OARs. Finally, patient selected criteria needs to be defined for using IC/IMRT in the future. IC/IMRT is potentially less invasive, and a more applicable treatment than IC/ISBT and ISBT. IC/IMRT is another alternative technique when IC/ISBT or ISBT is not feasible.