Positron Emission Tomography-Guided Bone Marrow-Sparing Radiation Therapy for Locoregionally Advanced Cervix Cancer: Final Results from the XXX Phase II/III Trial

Abstract Purpose To test effects of positron emission tomography (PET)-based bone marrow-sparing (BMS) image-guided intensity modulated radiation therapy (IMRT) on efficacy and toxicity for patients with locoregionally advanced cervical cancer. Methods and Materials In an international phase II/III trial, patients with stage IB-IVA cervical carcinoma were treated with either PET-based BMS-IG-IMRT (PET-BMS-IMRT group) or standard image-guided IMRT (IMRT group), with concurrent cisplatin (40 mg/m2 weekly), followed by brachytherapy. The phase II component non-randomly assigned patients to PET-BMS-IMRT or standard IMRT. The phase III trial randomized patients to PET-BMS-IMRT vs. IMRT, with a primary endpoint of progression-free survival (PFS) but was closed early for futility. Phase III patients were analyzed separately and in combination with phase II patients, comparing acute hematologic toxicity, cisplatin delivery, PFS, overall survival (OS), and patterns of failure. In a post-hoc exploratory analysis, we investigated the association between pretreatment absolute lymphocyte count (ALC) and OS. Results In total, 101 patients were enrolled on the phase II/III trial, including 29 enrolled in phase III (PET-BMS-IMRT group: 16; IMRT group: 13) before early closure. Median follow-up was 33 months for phase III patients and 39 months for all patients. PFS and OS at 5 years for all patients were 73.6% (95% CI 64.9%, 84.3%) and 84.0% (95% CI 76.0%, 92.9%), respectively. There were no differences in number of cisplatin cycles, OS, PFS, or patterns of failure between groups for the combined cohort. The incidence of acute grade ≥ 3 neutropenia was significantly lower in the PET-BMS-IMRT group compared to IMRT for randomized patients (19% versus 54%, c2 p=0.048) and in the combined cohort (13% vs. 35%, c2 p=0.01). Patients with pretreatment ALC ≤ 1.5 k/mL had non-significantly worse OS on multivariable analysis (HR 2.85, 95% CI 0.94, 8.62, adjusted p-value p=0.216), compared to patients with ALC > 1.5 k/mL. There was no difference in post-treatment ALC by treatment group. Conclusions PET-BMS-IMRT significantly reduced acute grade ≥ 3 neutropenia, but not treatment-related lymphopenia, compared to standard IMRT. We found no evidence that PET-BMS-IMRT impacted chemotherapy delivery or long-term outcomes, and weak evidence of an association between pre-treatment ALC and OS.