KEYNOTE-826 Data Supports Pembrolizumab in Patients With Cervical Cancer

Adding pembrolizumab (Keytruda) to chemotherapy with or without bevacizumab (Avastin) did not negatively affect health-related quality-of-life (QOL) in patients with recurrent, persistent, or metastatic cervical cancer, according to updated data from the patient-reported outcomes (PRO) analysis of the phase 3 KEYNOTE-826 trial (NCT03635567).

At a median follow-up of 22.0 months (interquartile range [IQR] 19.1-24.4) and at week 30, Quality-of-Life-Core 30 (QLQ-C30) completion was reached by 199 (69%) of 290 patients given pembrolizumab and 168 (57%) of 297 patients who received placebo. In these groups, compliance was 199 (94%) of 211 and 168 (90%) of 186, respectively.

From baseline to week 30, the least squares mean change in QLQ-C30 global health status (GHS)–QOL score was −0.3 points (95% CI, −3.1-2.6) with pembrolizumab and −1.3 points (−4.2-1.7) with placebo. There was a between-group difference in least squares mean change of 1.0 point (95% CI −2.7-4.7). Additionally, the median time to true deterioration in GHS–QOL in the pembrolizumab group was not reached (NR; 95% CI, 13.4 months-NR) compared with 12.9 months (95% CI, 6.6-NR) with placebo for a hazard ratio of 0.84 (95% CI, 0.65-1.09).

These data, along with previously reported efficacy and safety results from KEYNOTE-826, support the benefit of pembrolizumab and the value of immunotherapy in this patient population.

“Findings from the analysis of the overall improvement, stability, or deterioration in PRO scores showed that more patients had improved QLQ-C30 GHS-QOL in the pembrolizumab group than in the placebo group. Consistent with this finding, a slightly higher proportion of patients in the pembrolizumab group had improved EQ-5D-5L VAS [EuroQOL-5 dimension-5 level visual analogue scale] scores compared with the placebo group, and fewer patients had deteriorated scores,” the study authors wrote in findings published in Lancet Oncology.

The multicenter, randomized, phase 3 KEYNOTE-826 study aimed to assess the addition of pembrolizumab, an anti–PD-1 monoclonal antibody, to chemotherapy with or without bevacizumab across 151 cancer treatment centers in 19 countries.

Investigators previously reported that the study met its primary end points of improved overall survival (OS) and progression-free survival (PFS) with pembrolizumab vs placebo plus chemotherapy with or without bevacizumab. Additionally, treatment with pembrolizumab led to manageable toxicity in these patients with persistent, recurrent, or metastatic cervical cancer.

To be eligible for enrollment, patients must have been aged 18 years or older with persistent, recurrent, or metastatic cervical cancer who were not previously treated with systemic chemotherapy and were not amenable to curative treatment. Patients also were required to have an ECOG performance status of 0 or 1.

In the study, patients were randomly assigned by metastatic disease at diagnosis, planned bevacizumab use, and PD-L1 combined positive score (CPS) in a 1:1 fashion. They received either pembrolizumab 200 mg or intravenously infused placebo every 3 weeks for up to 35 cycles plus chemotherapy—consisting of paclitaxel 175 mg/m2 plus cisplatin 50 mg/m2 or carboplatin area under the curve 5 mg/mL per minute, intravenously—with or without bevacizumab 15 mg/kg every 3 weeks intravenously.

The PRO instruments were EORTC QLQ-C30, the EORTC cervical cancer module (QLQ-CX24), and the EQ-5D-5L VAS. Each of these were collected prior to treatment during cycles 1 through 14 and every other cycle thereafter.

Primary end points of the study were OS and PFS with a prespecified secondary end point of change from baseline in QLQ-C30 GHS0QOL. Other PRO analyses were protocol-specified exploratory end points.

Between November 20, 2018, and January 31, 2020, 883 patients were screened and 617 patients were randomly assigned. The pembrolizumab group consisted of 308 patients while the placebo group included 309. A total of 587 (95%) of 617 patients were given at least 1 dose of the study treatment, completed at least 1 post-baseline PRO assessment, and were included in the PRO analyses. This included 290 patients in the pembrolizumab arm and 297 patients in the placebo arm.

Among the patients included, 548 (89%) of 617 had a PD-L1 CPS of at least 1, including 273 in the pembrolizumab group and 275 in the placebo group. Baseline characteristics were similar between the treatment groups. In the pembrolizumab group, 45% of patients given pembrolizumab were non-White vs 39%of those given placebo; 21% and 15% were Asian, and 36% and 39% were Hispanic, respectively.

In the pembrolizumab group, 122 (42%) of 290 patients had improved GHS-QOL at any time during the study vs 85 (29%) of 297 in the placebo group (P = .0003).

Baseline mean QLQ-C30 GHS–QOL scores were 63.0 for the pembrolizumab group (SD, 23.3) and 66.3 for the placebo group (SD, 21.9), respectively. Through assessing the secondary end point, the least squares mean change in QLQ-C30 GHS–QOL score from baseline to week 30 in the all-comers group was −0.3 points in the pembrolizumab group (95% CI, –3.1-2.6) and –1.3 points (95% CI, –4.2-1.7) in the placebo group. Patients with PD-L1 CPS of at least 1 had a least squares mean change in QLQ-C30 GHS-QOL score from baseline to week 30 of 0.6 points (95% CI, –2.4-3.5) with pembrolizumab and –0.8 points (­95% CI, –3.9-2.4) with placebo for a between-group difference of 1.3 points (95% CI, –2.6-5.2; P = .50).

EQ-5D-5L VAS scores were generally similar in the study with baseline for both treatment groups up to week 51. The baseline least squares mean EQ-5D-5L VAS score for the pembrolizumab group was 70.5 (SD, 21.3) and 71.9 (SD, 20.2) in the placebo group. The least squares mean change in EQ-5D-5L VAS score from baseline to week 30 was 0.3 points (95% CI, –2.2- 2.8) with pembrolizumab and –1.5 points (95% CI, –4.1-1.1) with placebo. Moreover, the least squares mean difference in scores was 1.8 points (95% CI, –1.6-5.1; P = .29) between each treatment group.

For the median time to true deterioration in EQ-5D-5L VAS, it was NR (95% CI, 17.2 months-NR) in the pembrolizumab group vs 7.7 months (95% CI, 6.0-NR) in the placebo group for a hazard ratio of 0.75 (95% CI, 0.58-0.97; P = .027). The overall improvement or stability rate for EQ-5D-5L VAS scores was 78% in the pembrolizumab group and 72% in the placebo group (P = .033). In the pembrolizumab group, 43% of patients in the pembrolizumab group had improved scores vs 36% in the placebo group, 36% and 35% had stable scores, respectively, and 16% and 22% had deteriorated scores at any time during the study, respectively.

The improvement or stability rates for QLQ-CX24 subscales were similar between the treatment groups with a slightly higher proportion of patients in the pembrolizumab group having improved scores for the QLQ-CX24 cervical symptoms of lymphoedema, menopausal symptoms, and symptom experience, compared with the placebo group. Overall, complete PRO results for the population of patients with a PD-L1 CPS of at least 1 were generally similar to those for patients in the all-comers population.

“[W]e show that the significant improvements in OS and PFS with addition of pembrolizumab to the standard of care reported in the efficacy analysis of KEYNOTE-826 were not accompanied by deterioration in health-related QOL compared with placebo in the all-comer [ie, intention-to-treat] population and in patients with PD-L1 CPS of at least 1. Overall, findings from this PRO analysis provide further support for the use of pembrolizumab plus chemotherapy with or without bevacizumab as a new standard of care for persistent, recurrent, or metastatic cervical cancer,” concluded the study authors.

REFERENCE:

Monk BJ, Tewari KS, Dubot C, et al. Health-related quality of life with pembrolizumab or placebo plus chemotherapy with or without bevacizumab for persistent, recurrent, or metastatic cervical cancer (KEYNOTE-826): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2023;S1470-2045(23)00052-9. doi:10.1016/S1470-2045(23)00052-9