By Amishi Bajaj, MD, Northwestern Medicine Proton Center
This just in at ASTRO 2023: Vaginal cuff brachytherapy (VCB) has most commonly been completed over the course of 3-5 fractions but based on data from the SAVE trial out of the University of Utah Huntsman Cancer Institute, VCB may be safely performed in as few as 2 fractions with less acute urinary side effects and comparable gastrointestinal and vaginal side effects at one year with no significant difference in recurrence at 2.5-year median follow-up.
Some background on how we got to VCB: Adjuvant radiation therapy following surgical resection has traditionally been indicated for patients with early endometrial cancer for whom adverse risk factors are present. Such risk factors include patient age, extent of myometrial invasion, grade, histology (such as non-endometrioid histologies like serous or clear cell carcinoma), and lymphovascular space invasion (LVSI). Adjuvant external beam radiotherapy (EBRT) was shown to substantially decrease the risk of recurrence for patients with “high-intermediate risk” disease (defined differently by the two trials comparing post-op observation vs. adjuvant EBRT, GOG-99 and PORTEC-1, while still addressing the same group of risk factors [age, grade, myometrial invasion and LVSI]). With over 70% of recurrences occurring in the vaginal vault, PORTEC-2 then sought to compare adjuvant EBRT (46 Gy in 23 fractions) versus VCB (7 Gy in 3 fractions prescribed to 5 mm depth for patients receiving HDR). Though the recurrence rate in the pelvis was found to be higher for patients receiving VCB in PORTEC-2 (3.8% versus 0.5%, p=.02), the five-year vaginal recurrence rate was comparable (1.8% versus 1.6%, p=.74) and the rates of GI toxicity were considerably lower using VCB. The key to identifying the optimal candidacy for VCB, as always in oncology, is appropriate selection.
How do people prescribe VCB? Per the most recent NCCN guidelines as well as the American Brachytherapy Task Group Report/practice parameter on adjuvant VCB for early stage endometrial cancer (Harkenrider et al., 2017), there are a variety of VCB monotherapy regimens that may be considered standard, including 7 Gy x 3 fractions prescribed to 5 mm depth (as was utilized in the aforementioned PORTEC-2), 5.5 Gy x 4 fractions prescribed to 5 mm depth, and 6 Gy x 5 fractions prescribed to the surface. Use of these regimens acknowledges that dose prescription to a depth will ultimately result in higher dose to the mucosa, so per NCCN guidelines, dosing may be altered to optimize outcomes (and ensuring adequate coverage of the lymphatics deep to the mucosa) while minimizing vaginal mucosal toxicity.
The SAVE trial design: The SAVE trial is a prospective, randomized, multicenter trial with non-inferiority design randomizing women age ≥ 18 with ECOG performance status 0-2 and life expectancy exceeding two years s/p total hysterectomy/BSO with endometrioid, serous, clear cell, or sarcomatous histologies classified as: (1) stage II, grades 1-3, (2) stage IB, grades 1-3, or (3) stage IA, grade 1 with LVSI or stage IA grades 2-3. These patients were randomized in a 1:1 fashion to a standard-of-care approach (7 Gy in 3 fractions to 5 mm depth, 5-5.5 Gy in 5 fractions to 5 mm depth, or 6 Gy in 5 fractions prescribed to cylinder surface) to the experimental arm of 22 Gy in 2 fractions prescribed to cylinder surface. The primary endpoint was Global Health Status from the EORTC QLQ-C30 at one month post-treatment with secondary endpoints of patient-reported vaginal, bladder and bowel symptoms; physician-reported side effects by CTCAE version 5.0; recurrence patterns; HR-QOL; dose to organs at risk; and cost-effectiveness (ongoing).
How did the investigators come up with the dose used in the experimental arm? Existing literature (Young et al., 2015; Damast et al., 2015) as well as clinical experiences out of Yale School of Medicine/Smilow Cancer Hospital have demonstrated no isolated vaginal recurrence when prescribing 21-22 Gy in 2 fractions to the cylinder surface. PORTEC-2 demonstrated a clinically significant vaginal atrophy rate of nearly 40% when utilizing a prescription of 24.6 Gy in 3 fractions when measuring dose to cylinder surface. The goal of the SAVE trial was to offer comparable local control while offering the potential for decreased acute toxicity.
SAVE trial outcomes: Principal Investigator David Gaffney MD, presented these results: 108 patients were enrolled from five centers (with 54 patients in each arm) and included patients mostly with endometrioid adenocarcinoma (70%) followed by serous carcinoma (18%), with much less representation of undifferentiated carcinoma or other histologic subtypes. Most patients had FIGO stage IA or IB disease, with only two patients in the experimental arm and four patients in the control arm having stage II disease. While there was no significant difference in the patient population enrolled to each arm based on FIGO stage (p=0.08), there was a significant difference in the number of patients enrolled to each arm based on LVSI, with more patients in the control arm having LVSI (31.5% of patients enrolled on the control arm) as compared to the experimental arm (only 14.8% of patients enrolled on the experimental arm; p=0.035).
With regard to the primary endpoint, patients’ Global Health Status QLQ-C30 was non-inferior to standard of care at one month (p=0.01), at six months (p=0.0003), and at 12 months (p=0.02). In terms of secondary outcomes, there was no significant difference between the standard versus experimental treatments in terms of gastrointestinal symptoms or sexual activity; however, urinary toxicity was significantly different across all measured timepoints, with the greatest extent of difference noted at the one- to three-month timeframe. Of all patients noting any adverse events that were “possibly,” “probably,” or “definitely” related to VCB, there was no significant difference in all side effects with the exception of markedly decreased urinary events in the experimental arm as compared to the control arm (1 versus 14, p=0.0004).
So which patients could be treated with two fractions? Practice patterns may vary by radiation oncologist, but the SAVE trial demonstrates certain benefits for select patients and practice environments. These include patients with high anxiety at the prospect of being treated with a vaginal cylinder or those experiencing great discomfort upon cylinder placements. Further, for centers with fewer resources for administration of VCB, the ability to offer treatment in two fractions allows for more patients to be treated effectively without compromising resource availability.
Takeaways and Future Direction: The SAVE trial has demonstrated that performing VCB in two fractions is safe and well tolerated with a noted improvement in acute urinary toxicity. While local control has shown to be comparable with 2.3 year median follow-up, longer term follow-up data remains pending. Further, as stated previously, appropriate selection is critical: with fewer patients in the experimental arm having LVSI — a known strong risk factor for predicting local recurrence — this caveat must be kept in mind when interpreting the data. Per Principal Investigator, David Gaffney, MD, “A phase III expansion study is now ongoing, led by my colleague, Dr. Cristina DeCesaris, evaluating financial toxicity as the primary endpoint. We are very pleased to see at this time there is no compromise in local control, patient reported outcomes are non-inferior, and with fewer bladder adverse events in the 11 Gy x 2 arm. Now that we are entering the molecular era in endometrial cancer it will be of interest to see optimal vaginal cuff fraction schemes per molecular classification.”
Abstract 168 - Short Course Adjuvant Vaginal Cuff Brachytherapy (VCB) in Early Endometrial Cancer: Primary Endpoint Results of the SAVE Randomized Clinical Trial was presented on October 2, 2023, during the SS 13: GYN 2 – Improving Patient Centered Outcomes through Clinical Trials in Gynecological Cancers session at the ASTRO Annual Meeting.
