3510 - Local Recurrence Rates following Equivocal 3-month Post-Treatment PET after High Dose Rate Brachytherapy for Locally Advanced Cervical Cancer: An Updated Cohort

E. Harris¹, R. F. Shenker¹, J. W. C. Lee², S. J. Stephens¹, D. N. Ayala-Peacock³, and J. P. Chino¹; ¹Department of Radiation Oncology, Duke University Medical Center, Durham, NC, ²Duke University Medical Center, Durham, NC, ³Duke Cancer Institute, Raleigh-Durham, NC

Purpose/Objective(s): The standard treatment of locally advanced cervical cancer (LACC) includes chemoradiation (CRT) followed by brachytherapy (BT). Positron-emission tomography (PET) is often used to evaluate treatment response three months after completion of BT. PET response on radiology report, however, is often indeterminate within the high-risk clinical target volume (HRCTV). We aim to determine if equivocal PET response correlates with ongoing treatment response or persistent disease. This study provides an update of prior work presented at the ABS Annual Meeting in 2023. Materials/

Methods: This study was approved by our institutional IRB. Patients treated with CRT and high dose rate (HDR) BT for LACC at our institution from 2010–2020 were retrospectively reviewed. Those without 3-month post-treatment PET were not included. A complete response (CR) was defined as no concern for disease by treating physician and radiologist interpretation (RI). An equivocal response (ER) was defined by RI written as potential residual disease. Definite residual disease (RD) was indicated by radiology written interpretation as such. SUVmax for pre- and post-treatment PET was recorded. 5-year cervical control stratified by radiology read was determined by the Kaplan-Meier method; cohorts were compared by logrank test with a p-value of <0.05 set as the threshold for statistical significance.
Results: 98 patients were treated with CRT + HDR BT, of which 85 had post-treatment PET available for review. Median HRCTV D90 dose was 87.2 Gy (IQR 85.8 Gy – 89.4 Gy). Median HRCTV D98 was 78.0 Gy (IQR 76.5 Gy – 79.8 Gy). Median time from end of BT to PET was 3.3 months (IQR, 3.0 – 3.5). 5-year cervical control in all patients with post-treatment PET was 84.4% (95% CI 76.2 – 92.6). 76.5% of patients (n=65) were interpreted to have a CR, of which 2 experienced a local recurrence. 20% of patients (n=17) were interpreted to have an ER, of which 2 experienced a local recurrence. 3 patients were noted to have definite RD, each with clinically confirmed local recurrence. 5-year cervical control rate for those with a CR and ER by RI was 85.9% (95% CI 76.7 – 95.1%) and 94.1% (95% CI 82.9 – 100%), respectively. There was no difference in 5-year cervical control rate between CR and ER by RI (p= 0.490). 5-year cervical control rate for those with RD was 0%. 5-year cervical control rate of ER and CR compared to RD was significantly improved (p= <0.001 for both comparisons). Of those with an ER and with available pre- and post-treatment PET (n=12), the mean difference of SUVmax between pre- and post-treatment imaging was 10.5 (range 0.4 – 25.5).
Conclusion: Concern for residual disease on PET following definitive treatment for LACC is present in up to 1 of 5 patients. Local recurrences, however, in the HDR BT field are rare for patients with an equivocal post-treatment PET, with rates similar to those determined to have a CR. The majority of patients with equivocal post-treatment PET do not experience local failures and may be safely followed with serial imaging.