David M. O’Malley, MD, Department of Obstetrics and Gynecology at The Ohio State University College of Medicine and the director of the Division of Gynecologic Oncology at the OSUCCC – James speaks about the ESMO 2021 Abstract – LBA34 – Single-agent anti-PD-1 balstilimab or in combination with anti-CTLA-4 zalifrelimab for recurrent/metastatic (R/M) cervical cancer (CC): Preliminary results of two independent phase II trials.
Background:
The lack of second-line therapy for R/M CC remains a significant therapeutic need. In R/M CC, data from two phase 2 trials of balstilimab (bal) alone and in conjunction with zalifrelimab (zal) are presented.
Methods:
Patients received bal 3mg/kg q2w (NCT03104699) or bal 1mg/kg q6w (NCT03495882) as a single agent or in combination with zal 1mg/kg q6w for up to 2 years. The primary goal was objective response rates (ORR) as determined by independent review according to RECIST 1.1, with safety and DOR as secondary objectives.
Results:
In the bal and bal/zal, we treated 161 and 155 patients, respectively, with 160 and 143 patients having baseline quantifiable illness (modified ITT population). As per protocol, all pts had previously had platinum-based therapy for their first line. Squamous-cell carcinoma (SCC) was the most common histologic subtype (63 percent bal; 74 percent bal/zal), with adenocarcinoma/adenosquamous/other (AC) being prevalent. CPS 1% (62 percent bal; 55 percent bal/zal) was considered positive, whereas CPS 1% (26 percent bal; 25 percent bal/zal) was considered negative, and unknown (12 percent bal; 20 percent bal/zal) was considered unknown. The data on efficacy may be seen below.
In both studies, the treatment was well tolerated. Immune-related AEs occurred in 49 (30%) of patients in the bal trial and 50 (35%) in the bal/zal study (all grades), with severe (Grade 3+) AEs occurring in 13 (8.0%) and 15 (10.5%) of patients, respectively. Treatment discontinuation was seen in 22 points (13.7%) in bal and 15 points (10%) in bal/zal. The bal study had no treatment-related fatalities, whereas the bal/zal trial had two (nephritis; pneumonitis). There were no new safety signals discovered.
Conclusions:
Both single-agent bal and bal/zal are active and well-tolerated in R/M CC, according to these findings. Adding bal to zal enhanced ORR and DOR while increasing AEs only a little. Responses were more prevalent in PD-L1 + and SCC participants, although they were also evident in PD-L1- and AC participants. This is by far the greatest trial using checkpoint inhibitors in cervical cancer that has been published to date.
Clinical trial identification
NCT03104699, NCT03495882.