Results of the phase 1 NRG-GY017 trial show promise of atezolizumab as an immune primer in locally advanced cervical cancer.
Patients with locally advanced, node-positive cervical cancer who received immune priming with atezolizumab (Tecentriq) or concurrently with extended-field chemoradiation experienced promising progression-free survival (PFS) and few dose-limiting toxicities (DLTs), according to results of the randomized phase 1 NRG-GY017 trial (NCT03738228) that were presented at the The Society of Gynecologic Oncology (SGO) 2022 Annual Meeting on Women’s Cancer.1
Additionally, there was evidence of T-cell clonal expansion in the tumors and peripheral blood with the PD-L1 inhibitor.
At a median follow-up of 20 months, results showed an increase in peripheral blood T-cell receptor (TCR) clonal expansion and expansion of tumor-associated T-cell clones between the start of treatment and day 21 of chemoradiation with a priming dose of atezolizumab (P = .0001) and concurrent use of atezolizumab (P = .001).2
Additionally, patients who had higher pretreatment TCR diversity had a greater likelihood of complete pathologic response in on-treatment biopsy (P = .049).
“NRG-GY017 with concurrent atezolizumab or priming atezolizumab with chemoradiation was safe and very well tolerated. We think that the T-cell clonality expansion observed in tumors and the peripheral blood was irrespective of sequencing, but it’s quite interesting, and we hope to do more studies looking at this and correlating this [approach] with PFS and overall survival,” lead study author Jyoti Mayadev, MD, professor and assistant vice chair in the Department of Radiation Medicine and Applied Sciences at UC San Diego School of Medicine, said in a presentation during the meeting.
Cervical cancer remains a lethal disease with approximately 350,000 global deaths annually. Patients with node-positive disease continue to have poor prognosis, with recurrence rates up to 50% with standard chemoradiation.
Chemoimmunotherapy is approved for patients with metastatic cervical cancer, but its utility and optimal sequence with radiation in the locally advanced setting are unknown.
As such, investigators conducted the NRG-GY017 trial in which patients with stage IB-IVA disease with para-aortic nodal metastases or stage IIB-IVA disease with positive pelvic lymph nodes with or without para-aortic nodal metastases were randomized to 1 dose of priming atezolizumab plus 2 doses with chemoradiation (arm A; n = 19) or 3 doses of atezolizumab with concurrent chemoradiation (arm B; n = 17).
The primary objective was the evaluation of clonal expansion of TCR beta and tumor-associated TCR clones in the peripheral blood. Secondary objectives were safety and toxicity, with exploratory objectives of assessing the immune microenvironment and use of biopsy to identify predictive markers.
Tumor samples were collected before and during therapy, and peripheral blood was also collected. The immunoSEQ assay was used to evaluate TCR metrics.
DLTs were evaluated during and up to 30 days after chemoradiation.
A total of 40 patients were randomized; 36 patients who were evaluated received chemoradiation and at least 1 dose of atezolizumab.
The median age of patients was 47.5 years. The majority of patients were not Hispanic or Latino (n = 27; 75%), were White (n = 27; 75%), and had a performance status of 0 (n = 26; 72.2%). Most patients had squamous cell carcinoma (n = 28; 77.8%) and FIGO stage IIB disease (n = 22; 61.1%).
Additional results showed the clones that expanded in peripheral blood also expanded in the tumor in response to therapy. Notably, clonal expansion occurred most often in response to radiation vs atezolizumab, and most of the expanded clones were newly detected.
Furthermore, among 11 evaluable patients each in arms A and B, the complete response (CR) and CR/partial response rates were 45% and 82%,o and 27% and 36%, respectively.
The median PFS was not reached in either arm, with 4 events in arm A and 6 events in arm B. The 1-year PFS rate was approximately 72%, Mayadev said.
At data cutoff, 75% of patients had completed the study treatment. A total of 30 patients were evaluable for DLTs. None of the 16 patients in arm A experienced DLTs, and 3 of the 14 patients in arm B experienced DLTs (8%); these were immune-related colitis, non–immune related colitis, and thrombocytopenia with cisplatin delay of more than 2 weeks. Notably, the patient with immune-related colitis had the DLT resolved with medication.
Overall, 3 patients in arm A and 10 patients in arm B reported a grade 3 or higher treatment-related adverse effect, only 1 of which was immune related.
“The translational aspects of this study will also let us report more on how the tumor microenvironment will affect prognosis [and] functionality, and we also hope to look at imaging and PET data that was collected by the sites and the patients for long-term outcomes,” concluded Mayadev.