MR-guided SBRT boost for patients with locally advanced or recurrent gynecological cancers ineligible for brachytherapy: feasibility and early clinical experience

External beam radiotherapy (EBRT) with concurrent chemotherapy followed by a brachytherapy (BT) boost is the standard of care for most patients with locally advanced gynecological cancer [1]. On the other hand, the management of patients with non-metastasized recurrent gynecological cancer remains a challenge. If chemoradiotherapy followed by BT boost has not been performed before, this could be a treatment option to prevent extensive surgical resection, such as pelvic exenteration.

Even though the utilization of a sequential BT boost improves the outcomes in patients with locally advanced or recurrent gynecological cancer (LARGC) [2‐4], not every patient is suitable for a BT boost due to the extent or localization of the tumor, bone infiltration, infiltration of the pelvic wall or adjacent organs, or in multimorbid patients, who cannot undergo anesthesia.

As an alternative, the application of a stereotactic body radiotherapy (SBRT) boost delivered with cone beam computed tomography (CBCT) as image guidance and a linear accelerator (LINAC) has been performed in previous studies [5, 6]. However, due to the reduced soft tissue contrast, it remains challenging for the radiation oncologist to differentiate between tumor and normal tissue based on CBCT. As it is important to deliver high doses to the high-risk clinical target volume (HR-CTV) while sparing critical organs at risk (OAR), the use of high precision radiotherapy becomes indispensable.

Magnet resonance imaging (MRI) provides superior soft tissue contrast compared to CBCT. MRI-guided linear accelerators (MR-Linac) have incorporated an MRI scanner into a radiation therapy delivery system, allowing an improved visualization of gynecologic tumors and OAR [7, 8]. Online MR-guided radiotherapy (MRgRT) also provides the opportunity to adapt the treatment plan to interfractional anatomical changes and monitor intrafractional motion [9‐11], making this method a considerable option for patients with LARGC, who are not eligible for BT. In the present study, we investigated the feasibility and safety of an MRg-SBRT boost following EBRT.

With a median follow-up of 9 months (range, 8–19 months), a complete response (CR) was observed in 6 patients (60%), while local control was reported for 9 patients (90%). One of the 6 patients with CR developed inguinal lymph node metastases outside the radiation field within 12 months after the end of the MRg-SBRT boost. The patient received a salvage radiation treatment of the right inguinal lymphatic pathways with SIBs to lymph node metastases and is currently with no evidence of disease. Partial remission was seen in 2 patients, while a progression was reported in 2 patients: 1 patient developed distant metastases to mediastinal lymph nodes 7 months after the treatment, without locoregional recurrence; while 1 patient developed a local progression in the pelvis as well as distant metastasis within 6 months after therapy. These two patients were treated with salvage systemic therapy. Patient outcomes are described in Table 1.

Toxicities were reported according to the Common Terminology Criteria for Adverse Events (CTCAE) v. 5.0. At the end of pelvic EBRT, diarrhea CTC°I was reported in 3 patients (30.0%), proctitis CTC°I in 2 patients (20.0%), pollakisuria CTC°I in 2 patients (20.0%), dysuria CTC°I-II in 6 patients (60.0%), nycturia CTC°I-II in 5 patients (50.0%), urinary urgency CTC°I-II in 4 patients (40.0%), and dermatitis CTC°I-II in 2 patients (20.0%). No adverse events ≥ grade III were observed. There were no worsening of aforementioned acute toxicities after MRg-SBRT boost.

More than 3 months after MRg-SBRT boost, we observed CTC°I nycturia (40%), CTC°I dysuria (30%), and CTC°I urinary urgency (30%) as the most common side effects. Toxicities are summarized in Table 3.

In the present analysis we reported the early results of patients LARGC, who were ineligible for a sequential BT treatment and alternatively treated with an MRg-SBRT boost.

To date there are limited studies, which investigated the role of an EBRT boost in patients ineligible for BT. Barraclough et al. analyzed 44 patients with cervical cancer, who were not suitable for BT and therefore received a conventionally fractionated EBRT boost subsequently to pelvic irradiation. The EBRT boost was delivered with a total dose of 15.0–25.0 Gy in 8–10 fractions, resulting in a cumulative dose of 54.0–70.0 Gy. Even though late grade 3 toxicity was only seen in 2% of patients, recurrent disease was reported in 48% of the patients after a 2.3 year follow-up. The high recurrence rate was due to insufficient dose coverage in the large boost volume (median: 228 ccm) and dose limitations of the rectum, bladder and small bowel [16].

Other retrospective studies reported their experience on using SBRT to mimic BT in patients with LARGC, ineligible for BT. SBRT could be advantageous over conventionally fractionated EBRT, because of its ability to deliver higher dose to the tumor, and reduce the dose exposure of OAR at the same time. Guckenberger et al. evaluated the outcome of an SBRT boost after 50.0 Gy whole pelvic radiation in patients with LARGC. The SBRT boost was delivered with a total dose of 15.0 Gy in 3 fractions. There were no OAR dose exposure data regarding D0.1 ccm, D1ccm, and D2 ccm of the rectum. It was assumed the parts of the rectum receiving a higher dose in SBRT, were also fully exposed to EBRT of the pelvis. Despite the excellent local control rate of 81% after 3 years, grade ≥ 3 late sequelae were reported in 25%. Among 16 patients, 10.5% patients developed grade 4 intestinovaginal fistulae and 5.3% patients developed grade 4 small bowel ileus after a median follow-up of 22 months. Patients, who developed grade 4 intestino-vaginal fistulae, received a higher dose to the rectum (D_max 80–100 Gy EQD2₃). Other factors that caused the high rate of toxicities were large target volumes (median PTV volume: 92 ccm), and the invasion of the pelvic wall in most of the patients [17].

Another more recent study from Albuquerque et al. analyzed 15 patients with locally advanced cervical cancer (LACC), who were unable to receive BT and were treated with a SBRT boost. A total dose of 28.0 Gy in 4 fractions was delivered to the PTV. The median PTV volume was 139 ccm, which was larger than the median PTV volume in the present study. The rectum received a median total dose of 90.6 Gy (D2ccm, EQD2₃), which was much higher compared to our study. Similar to Guckenberger et al., they also reported 26.7% grade ≥ 3 toxicities, mostly rectal ulcers or rectovaginal fistulae related to patients with very large tumors. Two of these patients (13%) died from sepsis and bleeding after refusing colostomy [18]. Comparable rates of grade 2–3 toxicities were reported by Kubicek et al. despite a smaller boost volume [6].

Although the current National Comprehensive Cancer Network (NCCN) Guidelines for the treatment of cervical cancer do not recommend SBRT as an appropriate routine alternative to BT [19], SBRT boost might remain a salvage option in patients unable to receive BT. In the current study, a combination of pelvic EBRT and MRg-SBRT boost was well tolerated with encouraging early results. Due to superior soft tissue contrast, the use of MR-guidance allowed for better visualization of normal and tumor tissue compared to CBCT. Gynecological cancers are well known to have large inter-fraction movements due to different filling or surrounding organs [20]. The sequential MRIs during the SBRT boost enables to adjust not only the tumor and OAR volumes based on daily anatomy, but also to capture inter-/ intrafraction movement, and to re-optimize the treatment plan accordingly. Hence, we were able to maximize dose to the target volume, while sparing the OARs [21]. However, the authors would like to emphasize that MRg-SBRT should never replace the treatment of choice–which is and will remain brachytherapy.

Regarding the tumor volume, we reported smaller median PTVs than in the previous studies from Guckenberger et al. and Albuquerque et al. [17, 18], nonetheless the PTVs are comparable to other similar studies [6, 22]. In order to preserve the OARs and prevent high rates of toxicity, the median cumulative dose to the PTV was lower than the recommendation in the EMBRACE II protocol [14]. It has been reported in a previous study, that the D1ccm, as well as D2ccm of the rectal wall are predictive for chronic rectal toxicity, moreover applying traditional dose multimodal concepts of EBRT and BT resulted in acceptable late toxicities of 10–15% [23]. The results from the prospective multicenter EMBRACE study showed that D2ccm of the rectum ≤ 65.0 Gy was correlated with less frequent rectal morbidity [24]. Therefore, we followed all classical constraints from brachytherapy (EMBRACE II protocol) regarding rectum, bladder, sigmoid and bowel. More than 3 months after the MRg-SBRT, we found 30% grade 1 dysuria, 40% grade 1 nycturia, 30% urinary urgency, but no grade ≥ 2 toxicities.

With an early median follow-up of 9 months, we observed a local control rate of 90% with 6 patients with CR, 2 patients with PR, and 2 patients with PD. The patient with a locoregional progression also developed distant metastases 6 months after the therapy. She had the largest tumor volume among all patients (PTV = 131.35 ccm), with infiltration of the pelvic wall and other adjacent organs. This finding is in accordance with Ijaz et al., who reported extension to the pelvic wall as a negative prognostic factor for survival [25].

Some other retrospective studies analyzed the utilization of a SBRT boost in patients ineligible for BT. There were heterogeneous results in terms of toxicity and outcome. Unknown tumor volume, different radiation techniques and fractionation schedules make it difficult to compare them to our results [5, 6, 17, 18, 22, 26‐29]. A summary of the current literature is described in Table 4.

The current study inherits several limitations. A longer follow-up is necessary, to rule out high grade chronic toxicity and to obtain long-term outcomes. Our cohort was small and heterogeneous, so that a comparison between the efficacy of a SBRT boost and BT is difficult. However, as high-dose levels to the HR-CTV are very important for local control rates [30], BT will always remain superior and the standard of care [21]. Therefore, MRg-SBRT boost may serve as a backup solution in selected patients, who are ineligible for BT. The current study revealed that MRg-SBRT seems a safe therapeutic option and could be favorable over CBCT-guided SBRT.

These early results report the feasibility of an MRg-SBRT boost approach in patients with LARGC who were not candidates for BT. When classical BT-OAR constraints are followed, the therapy was well tolerated. However, long-term follow-up is needed to validate the present hypothesis.