Integrating CTLA-4/PD-1 blockade into cervical cancer management: results of COMPASSION-16

Summary

Although there is anti-tumor efficacy of dual CTLA-4/PD-1 blockade in advanced/recurrent cervical cancer, it is unclear whether combination with chemotherapy is synergistic. In COMPASSION-16, Wu et al. demonstrated improved survival outcomes of cadolinimab plus chemotherapy compared to chemotherapy alone for first-line systemic therapy for advanced/recurrent cervical cancer, suggesting a potential role of bispecific CTLA-4/PD-1 inhibitors in the frontline setting.

Cervical cancer continues to be a major global health issue with an annual incidence and mortality of 660,000 cases and 350,000 deaths worldwide [1]. Management of advanced/recurrent cervical cancer has rapidly evolved over the past decade. Prior to the advent of immune checkpoint inhibitors, the phase III Gynecologic Oncology Group (GOG) cooperative group 240 trial established chemotherapy plus bevacizumab as the standard first-line systemic therapy for advanced or recurrent cervical cancer [2]. In GOG-240, chemotherapy (cisplatin / paclitaxel or topotecan / paclitaxel) plus bevacizumab was found to have a significantly improved median overall survival (OS) compared to chemotherapy alone (24.5 months vs 16.8 months, respectively) [2, 3]. Demonstrating durable and impressive responses, PD-1 inhibitors have been increasingly utilized in the management of multiple solid tumors. In a recent phase III trial, the KEYNOTE-826 investigators evaluated the efficacy of chemotherapy (carboplatin / paclitaxel +/− bevacizumab) plus pembrolizumab versus chemotherapy alone as first-line systemic therapy for advanced/recurrent cervical cancer [4, 5]. Among PD-L1+ tumors, the addition of pembrolizumab to chemotherapy demonstrated improved median progression-free survival (PFS) (10.4 vs 8.2 months) and OS (28.6 vs 16.5 months) compared to chemotherapy alone [4, 5]. Thus, the current standard for first-line systemic therapy for advanced/recurrent cervical cancer is platinum/taxane chemotherapy plus bevacizumab with the addition for pembrolizumab for PD-L1 positive tumors. Dual CTLA-4 and PD-1 inhibitors have demonstrated anti-tumor activity as second-line systemic therapy and there is evidence that this strategy may have enhanced efficacy in chemotherapy-naïve patients [6–8]. However, it was unclear whether the addition of dual CTLA-4/PD-1 blockade to chemotherapy would improve treatment efficacy. Therefore in COMPASSION-16 (NCT04982237), Wu and colleagues sought to evaluate the efficacy and safety of chemotherapy with or without cadolinimab (an anti-PD-1 and CTLA-4 bispecific monoclonal antibody) as first-line systemic therapy for advanced/recurrent cervical cancer [9]. These results were published in Lancet and will be reviewed here [9].

COMPASSION-16 was a multicenter phase III trial in China that randomized 445 cervical cancer patients 1:1 to cadolinimab or placebo plus platinum/taxane chemotherapy with or without bevacizumab for 6 cycles followed by maintenance therapy (cadolinimab or placebo with or without bevacizumab) for up to 2 years [9]. The dual primary endpoint was PFS and OS. Key eligibility criteria included 1) persistent, recurrent or stage IVB disease 2) squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma histology 3) not amenable to curative surgery or concurrent chemoradiation and 4) no prior systemic therapy for persistent, recurrent or metastatic disease. Of note, PD-L1 positivity was not requirement.

The cadolinimab and placebo groups were balanced in terms of baseline characteristics. The majority of the patient population had squamous cell carcinoma histology (83%), had not received prior chemoradiation (52%), had a PD-L1 Combined Positivity Score (CPS) ≥1 (70%), and received concurrent bevacizumab during the trial (60%). With a median follow-up of 25.6 months, median PFS was significantly longer in the cadolinimab group compared to the placebo group (12.7 vs 8.1 months; HR 0.62, 95% CI 0.49 – 0.80, p<0.001). Among PD-L1 positive tumors, the cadolinimab group had greater PFS compared to the placebo group (13.5 vs 8.3 months; HR 0.62, 95% CI 0.46 – 0.83). Despite no statistical significance among PD-L1 negative tumors, there was a trend in median PFS favoring in the cadolinimab group (12 vs 8.2 months; HR 0.73, 95% CI 0.46 – 1.17). Although PFS was similar among patients with concomitant bevacizumab use (15.1 vs 11.5 months; HR 0.81, 95% CI 0.58 – 1.13), PFS was more favorable for the cadolinimab group among those without concomitant bevacizumab use (11.7 vs 6.9 months; HR 0.46, 95% CI 0.32 – 0.66).

For OS, the median OS was greater in cadolinimab group compared to the placebo group (not reached vs 22.8 months; HR 0.64, 95% CI 0.48 – 0.86, p = 0.0011). Similar to PFS in the PD-L1 positive tumor population, cadolinimab demonstrated superior median OS (not reached vs 22.7 months; HR 0.69, 95% CI 0.49 – 0.97). For the PD-L1 negative tumor population, there was a trend in improved median OS (not reached vs 25.3 months; HR 0.77, 95% CI 0.44 – 1.34). Like PFS, median OS was similar among the subpopulation with concomitant bevacizumab use (not reached in both; HR 0.84, 95% CI 0.56 – 1.26) but improved in the subpopulation without bevacizumab use (28.2 vs 15.1 months; HR 0.50, 95% CI 0.33 – 0.75). Objective response rate was higher in the cadolinimab group (83% vs 69%, respectively) along with a higher proportion of complete responses (36% vs 23%, respectively). Grade≥3 treatment-related adverse events were higher in the cadolinimab group (85% vs 80%, respectively) with the most common being hematologic (e.g. decreased neutrophil count, decreased white blood cell count, and anemia). Grade≥3 immune-related adverse events occurred in 10% of the cadolinimab group with the most common being hypopituitarism (0.9%), hepatitis (0.9%), and diabetic ketoacidosis (0.9%). Treatment discontinuation of any trial agent related to adverse events occurred more in the cadolinimab group (28% vs 11%, respectively).

The COMPASSION-16 investigators should be applauded for work that demonstrates a clear survival benefit of cadolinimab plus chemotherapy over chemotherapy alone for first-line systemic therapy for advanced/recurrent cervical cancer while maintaining an acceptable safety profile [9]. As discussed earlier, there has been a significant shift in treatment paradigm for cervical cancer without prior systemic therapy with the recent publication of KEYNOTE-826 [4, 5]. It is therefore unclear whether chemotherapy plus cadolinimab will provide superior treatment efficacy compared to chemotherapy plus PD-1 inhibition (e.g. pembrolizumab). Of note, KEYNOTE-826 demonstrated that the addition of pembrolizumab to chemotherapy mainly provided PFS and OS benefit for PD-L1 positive tumors (CPS≥1) [4, 5]. In contrast, the impact of PD-L1 status on survival benefit when cadolinimab was utilized was less clear in COMPASSION-16 [9]. Although there was no statistically significant difference in PFS or OS between treatment arms among those with tumors with CPS<1, the median PFS (12 months) and OS (not reached) achieved similar absolute values as those with tumors with CPS≥1 [9], and the less evident efficacy benefit is likely due to the small number of PD-L1 negative tumors. Additionally, the impact of cadolinimab appeared to be most influential on PFS and OS when concomitant bevacizumab was not utilized [9]. In contrast, KEYNOTE-826 demonstrated chemotherapy plus pembrolizumab demonstrated improved PFS when concomitant bevacizumab was used but did not demonstrate improvement in PFS when bevacizumab was omitted [4]. Therefore, cadolinimab may play a greater role when added to chemotherapy in situations when bevacizumab use is contraindicated.

With the incorporation of immune checkpoint inhibitors in both the frontline and recurrent setting for cervical cancer, data regarding treatment sequencing is currently lacking, and will become increasingly important to maximize overall survival. O’Malley et al. showed CTLA-4 and PD-1 inhibitors (zalifrelimab and balstilimab) had a 25.6% response rate as second-line treatment of recurrent cervical cancer [8]. In a Phase I/II trial (CheckMate-358), Oaknin et al. showed that dual CTLA-4 and PD-1 blockade (ipilimimab and nivolumab) in chemotherapy- and immunotherapy-naïve recurrent cervical cancer patients produced higher response rates and sustained anti-tumor benefit when compared with patients who had prior systemic treatment [6]. This suggests that there may be a role for combined immune checkpoint blockade before chemotherapy. Currently, there is an ongoing trial [NCT05475171] evaluating the use of a bispecific CTLA-4/PD-1 inhibitor (lorigerlimab) prior to physician’s choice systemic therapy (platinum/taxane chemotherapy +/− pembrolizumab +/− bevacizumab) in chemotherapy-naïve advanced or recurrent cervical cancer. This trial seeks to evaluate the use of dual checkpoint blockade as a sequencing triage strategy to improve OS as there may be a subpopulation that greatly benefits from immune checkpoint inhibitors. Additionally, this strategy may significantly delay or potentially avoid use of further line systemic chemotherapy.

The results of COMPASSION-16 provide an early signal that bispecific CTLA-4/PD-1 inhibition may play a role in combination with first-line chemotherapy in advanced/recurrent cervical cancer. Future studies should evaluate the role cadolinimab and other bispecific CTLA-4/PD-1 inhibitors and whether it more provide greater OS benefit over chemotherapy plus PD-1 inhibitors or CTLA-4/PD-1 blockade alone.