Advertisement

EMPOWER CERVICAL-1: Effects of cemiplimab versus chemotherapy on patient-reported quality of life, functioning and symptoms among women with recurrent cervical cancer

Open AccessPublished:July 31, 2022DOI:https://doi.org/10.1016/j.ejca.2022.03.016

      Highlights

      • This study examined patient-reported outcomes for cervical cancer treatments.
      • Patient-reported outcomes generally were stable/improved with cemiplimab and worsened with chemotherapy.
      • Quality of life and physical functioning significantly favoured cemiplimab in patients with squamous cell carcinoma.
      • Many other functioning and symptom scales also favoured cemiplimab versus chemotherapy.

      Abstract

      Background

      In a phase III, randomised, active-controlled study (EMPOWER-Cervical 1/GOG-3016/ENGOT-cx9; R2810-ONC-1676; NCT03257267) and cemiplimab significantly improved survival versus investigator's choice of chemotherapy among patients with recurrent cervical cancer who had progressed on platinum-based therapy. Here we report patient-reported outcomes in this pivotal study.

      Methods

      Patients were randomised 1:1 to open-label cemiplimab (350 mg intravenously every 3 weeks) or investigator's choice of chemotherapy in 6-week cycles. Patients completed the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 during cycles 1–16. Least-squares mean changes from baseline in global health status (GHS)/quality of life (QoL) and physical functioning (PF) were secondary end-points in the statistical hierarchy.

      Results

      Of 608 patients (304/arm), 77.8% patients had squamous cell carcinoma and 22.2% patients had adenocarcinoma. Questionnaire completion rates were ∼90% throughout. In the squamous cell carcinoma population, overall between-group differences statistically significantly favoured cemiplimab in GHS/QoL (8.49; 95% confidence interval [CI]: 3.77–13.21; P = 0.0003) and PF (8.35; 95% CI: 4.08–12.62; P < 0.0001). Treatment differences favoured cemiplimab in both histologic populations by cycle 2. Overall changes from baseline in most functioning and symptom scales favoured cemiplimab, with clinically meaningful treatment differences in role functioning, appetite loss and pain in both populations. The sensitivity analyses, responder analyses and time to definitive deterioration favoured cemiplimab in both populations.

      Conclusions

      Cemiplimab conferred favourable differences in GHS/QoL and PF compared with chemotherapy among patients with recurrent cervical cancer, with benefits in PF by cycle 2, and clinically meaningful differences favouring cemiplimab in role functioning, appetite loss, and pain.

      Graphical abstract

      Keywords

      1. Introduction

      Cervical cancer is a leading cause of cancer-related death among women [
      • Sung H.
      • Ferlay J.
      • Siegel R.L.
      • Laversanne M.
      • Soerjomataram I.
      • Jemal A.
      • et al.
      Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.
      ]. Survival for patients with cervical cancer has improved in high-income countries with better prevention and detection, as well as more effective treatments [
      • American Society of Clinical Oncology
      Cervical cancer: Statistics.
      ]. Worldwide, there is no current standard of care for recurrent or metastatic cervical cancer after first-line treatment failure, and significant impairments of quality of life (QoL), functioning and symptoms remain [
      • Corkum M.T.
      • Shaddick H.
      • Jewlal E.
      • Patil N.
      • Leung E.
      • Sugimoto A.
      • et al.
      When Pap testing fails to prevent cervix cancer: a qualitative study of the experience of screened women under 50 with advanced cervix cancer in Canada.
      ,
      • Herzog T.J.
      • Wright J.D.
      The impact of cervical cancer on quality of life--the components and means for management.
      ]. In patients with advanced cancer, physical well-being is significantly associated with overall survival, with a 3.7% decrease in risk of death for each unit of improvement in physical well-being [
      • Chase D.M.
      • Huang H.Q.
      • Wenzel L.
      • Cella D.
      • McQuellon R.
      • Long H.J.
      • et al.
      Quality of life and survival in advanced cervical cancer: a Gynecologic Oncology Group study.
      ].
      Patient-reported outcomes (PROs) are used by clinicians and patients to assess treatment choices and shape guidelines and by regulatory authorities and policy makers to evaluate risk–benefit profiles of medicines [
      • Addario B.
      • Geissler J.
      • Horn M.K.
      • Krebs L.U.
      • Maskens D.
      • Oliver K.
      • et al.
      Including the patient voice in the development and implementation of patient-reported outcomes in cancer clinical trials.
      ]. Measuring what matters to patients is of critical importance in cancer clinical trials [
      • Bhatnagar V.
      • Hudgens S.
      • Piault-Louis E.
      • Jones L.
      • Beaver J.A.
      • Lyerly H.K.
      • et al.
      Patient-reported outcomes in oncology clinical trials: stakeholder perspectives from the accelerating anticancer agent development and validation workshop 2019.
      ] and developing new treatments that improve survival must be contextualised with their effects on PROs.
      Cemiplimab is a human immunoglobulin G4 monoclonal antibody to the programmed cell death-1 receptor. In an international, phase III, randomised, active-controlled study of patients with recurrent or metastatic cervical cancer after first-line platinum-based chemotherapy (EMPOWER-Cervical 1/GOG-3016/ENGOT-cx9; R2810-ONC-1676; NCT03257267), cemiplimab significantly improved overall survival compared with investigator's choice of chemotherapy. This report describes PRO results in this pivotal study.

      2. Methods

      2.1 Study design

      The study design was described previously in full [
      • Tewari K.S.
      • Monk B.J.
      • Vergote I.
      • Miller A.
      • de Melo A.C.
      • Kim H.S.
      • et al.
      Survival with cemiplimab in recurrent cervical cancer.
      ]. Briefly, 608 patients from approximately 100 sites globally were randomised 1:1 to either the experimental or the control group and received open-label study treatment for up to 96 weeks in 6-week treatment cycles. In the experimental group, cemiplimab 350 mg was administered intravenously every 3 weeks. In the control group, the investigator chose chemotherapy from the following: antifolate, topoisomerase 1 inhibitor, nucleoside analogue and vinca alkaloid (Supplementary Fig. 1).
      All patients had relapsed or metastatic cervical cancer (squamous cell carcinoma [SCC] or adenocarcinoma/adenosquamous carcinoma [AC]) that had progressed after first-line platinum-based chemotherapy. Randomisation was stratified by histology (SCC versus AC), geographic region (North America versus Asia versus rest of world), prior use of bevacizumab (yes versus no) and Eastern Cooperative Oncology Group (ECOG) performance status (0 versus 1). All patients provided written informed consent. An institutional review board or ethics committee reviewed and approved the study protocol and informed consent form for each site.

      2.2 PROs

      Patients completed the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30) at baseline and on day 1 of each 6-week treatment cycle, up to cycle 16 or disease progression, and then at two post-treatment follow-up visits, with a recall period for each question of ‘during the past week.’ Scores on each EORTC QLQ-C30 scale range from 0 to 100; higher scores for global health status (GHS)/QoL and functioning scales indicate better health status and function, and higher scores for symptom scales indicate more symptoms [
      • Fayers P.M.
      • Aaronson N.K.
      • Bjordal K.
      • Groenvold M.
      • Curran D.
      • Bottomley A.
      • et al.
      The EORTC QLQ-C30 scoring manual.
      ]. A 10-point threshold for clinically meaningful differences [
      • King M.T.
      The interpretation of scores from the EORTC quality of life questionnaire QLQ-C30.
      ,
      • Osoba D.
      • Rodrigues G.
      • Myles J.
      • Zee B.
      • Pater J.
      Interpreting the significance of changes in health-related quality-of-life scores.
      ,
      • Cocks K.
      • King M.T.
      • Velikova G.
      • Fayers P.M.
      • Brown J.M.
      Quality, interpretation and presentation of European Organisation for Research and Treatment of Cancer quality of life questionnaire core 30 data in randomised controlled trials.
      ] was used to assess changes from baseline (within patient or within group) and differences between treatment groups.

      2.3 Statistical analysis

      Analyses were performed on the full analysis set, defined as all randomised patients, using SAS version 9.4 (SAS Institute, Cary, NC, USA). Longitudinal analyses included patients in the full analysis set with EORTC QLQ-C30 assessments at baseline and at least once after the baseline visit.
      Instrument completion rate at each timepoint was summarised. Mixed models for repeated measures (MMRM) were used to estimate least-squares (LS) mean changes from baseline. The following covariates were included as fixed effects: treatment arm, timepoint, baseline PRO score, geographic region, histology (for the overall population only), interaction of baseline PRO score and timepoint, and interaction of treatment arm and timepoint. In each population (SCC, AC and overall), MMRM analyses did not include cycles when the sample size dropped below 10 patients in either treatment group.
      The statistical hierarchy (Supplementary Table 1) included MMRM analysis of GHS/QoL and physical functioning, in the SCC and overall populations, for overall change from baseline and change from baseline to cycle 2 (week 6). In another advanced cervical cancer trial, median times to response and progression were approximately 2 months [
      • Chung H.C.
      • Ros W.
      • Delord J.P.
      • Perets R.
      • Italiano A.
      • Shapira-Frommer R.
      • et al.
      Efficacy and safety of pembrolizumab in previously treated advanced cervical cancer: results from the phase II KEYNOTE-158 study.
      ], suggesting that cycle 2 was the best opportunity to understand clinical benefit in this study. Secondary end-points in the statistical hierarchy were assessed for statistically significant differences at a one-sided P value of 0.025, with multiplicity adjustment. After the first end-point in the hierarchy failed this test, subsequent comparisons were not tested formally. End-points not in the statistical hierarchy were tested at a two-sided P value of 0.05, without multiplicity adjustment. P values after the break of hierarchy were nominal. Other statistical methods are described in the Supplementary Methods.

      3. Results

      3.1 Baseline characteristics and disposition

      Of 608 patients randomised (304 in each group), 300 received at least one dose of cemiplimab and 290 received at least one dose of chemotherapy. The treatment groups were similar at baseline (Table 1). Most patients had SCC histology (77.8%) and metastatic disease (94.4%).
      Table 1Patient characteristics (reprinted from Tewari et al. [
      • Tewari K.S.
      • Monk B.J.
      • Vergote I.
      • Miller A.
      • de Melo A.C.
      • Kim H.S.
      • et al.
      Survival with cemiplimab in recurrent cervical cancer.
      ]).
      Cemiplimab (n = 304)Chemotherapy (n = 304)Overall (N = 608)
      Age, years
       Mean (SD)51.1 (11.59)51.2 (11.77)51.1 (11.67)
       Range22–8124–8722–87
       ≥65–<75, n (%)30 (9.9)29 (9.5)59 (9.7)
       ≥75, n (%)5 (1.6)11 (3.6)16 (2.6)
      Race, n (%)
       White193 (63.5)192 (63.2)385 (63.3)
       Asian88 (28.9)88 (28.9)176 (28.9)
       Black or African American9 (3.0)12 (3.9)21 (3.5)
       Other or not reported14 (4.6)12 (3.9)26 (4.3)
      Geographic region, n (%)
       North America32 (10.5)34 (11.2)66 (10.9)
       Asia83 (27.3)83 (27.3)166 (27.3)
       Rest of world189 (62.2)187 (61.5)376 (61.8)
      ECOG performance status, n (%)
       0142 (46.7)141 (46.4)283 (46.5)
       1162 (53.3)163 (53.6)325 (53.5)
      Histology/cytology, n (%)
       SCC240 (78.9)233 (76.6)473 (77.8)
       Adenocarcinoma54 (17.8)62 (20.4)116 (19.1)
       Adenosquamous cell carcinoma10 (3.3)9 (3.0)19 (3.1)
      Extent of disease, n (%)
       Metastatic284 (93.4)290 (95.4)574 (94.4)
       Recurrent/persistent20 (6.6)14 (4.6)34 (5.6)
      ECOG, Eastern Cooperative Oncology Group; SCC, squamous cell carcinoma; SD, standard deviation.
      At baseline, study participants had a low-to-moderate disease burden for GHS/QoL and functioning scores compared with normative values in the general population (Supplementary Fig. 2) [
      • Hinz A.
      • Singer S.
      • Brahler E.
      European reference values for the quality of life questionnaire EORTC QLQ-C30: results of a German investigation and a summarizing analysis of six European general population normative studies.
      ]. Baseline GHS/QoL and functioning scores among study participants were slightly lower than reference values for patients with any cancer or cervical cancer (where available) [
      • Scott N.W.
      • Fayers P.
      • Aaronson N.K.
      • Bottomley A.
      • de Graeff A.
      • Groenvold M.
      • et al.
      EORTC QLQ-C30 reference values.
      ]. Baseline symptom scale scores in the SCC population were also generally similar between the treatment groups, except for higher pain scores with cemiplimab, and were consistent with reference values for patients with any cancer or cervical cancer.
      Patients progressively discontinued the study due to disease progression, death or other reasons (e.g. lost to follow-up, patient or physician decision, withdrawal of consent). The number of expected PRO assessments, therefore, decreased during the study for both the SCC population (Supplementary Fig. 3A) and the overall population (Supplementary Fig. 3B). Among patients still on study, >95% completed at least one PRO scale at baseline and approximately 90% completed at least one PRO scale at each subsequent assessment.
      Results were evaluable for ≥10 patients in each treatment group through cycle 7 in the SCC population (at cycle 8: cemiplimab, n = 49; chemotherapy, n = 8), through cycle 8 in the overall population (at cycle 9: cemiplimab, n = 54; chemotherapy, n = 8) and through cycle 4 in the AC population (at cycle 5: cemiplimab, n = 23; chemotherapy, n = 6). Due to the limited PRO sample and follow-up for the AC population, results are presented herein only for the SCC and overall populations.

      3.2 GHS/QoL

      MMRM results for GHS/QoL showed that at each post-baseline visit LS mean scores were generally maintained or improved with cemiplimab, while those for chemotherapy generally worsened (95% confidence intervals [CIs] did not include null) at cycles 2–4, both in the SCC population (Fig. 1A) and in the overall population (Fig. 1B). The overall LS mean difference between treatment groups in GHS/QoL change from baseline was statistically significant in favour of cemiplimab in the SCC population (8.49; 95% CI: 3.77–13.21; P = 0.0003) (Fig. 2A). The LS mean difference between treatment groups in change from baseline for GHS/QoL at cycle 2 was 4.14 (95% CI: −0.21, 8.50; P = 0.031) in the SCC population, but the P value was >0.025; per the statistical hierarchy (Supplementary Table 1), all subsequent comparisons were nominal. LS mean changes from baseline favoured cemiplimab in the overall population (7.81; 95% CI: 3.30–12.33; P = 0.0004) (Fig. 2B). The LS mean difference between the cemiplimab and chemotherapy groups in change from baseline for GHS/QoL at cycle 2 in the overall population was 3.25 (95% CI: −0.54, 7.03; P = 0.046).
      Fig. 1
      Fig. 1Change from baseline to each visit for pre-specified secondary endpoints. GHS/QoL in (A) the SCC population and (B) the overall population. Physical functioning in (C) the SCC population and (D) the overall population. Scheduled visits with <10 patients in either arm, unscheduled visits and off-treatment visits were not included in the analysis. C, cycle; CI, confidence interval; D, day; GHS, global health status; LS, least squares; QoL, quality of life; SCC, squamous cell carcinoma.
      Fig. 2
      Fig. 2Overall changes from baseline and differences between treatment groups. GHS/QoL and functioning scales in (A) the SCC population and (B) the overall population. Symptoms in (C) the SCC population and (D) the overall population. CI, confidence interval; GHS, global health status; LS, least squares; QoL, quality of life; SCC, squamous cell carcinoma; SE, standard error.
      Sensitivity analysis using a pattern-mixture model (PMM) showed clinically meaningful (i.e. ≥10 points) differences favouring cemiplimab across multiple delta values in both populations (Supplementary Fig. 4A and 4B). The proportion of responders (i.e. patients who improved by ≥ 10 points from baseline) was higher and the proportion of non-responders (i.e. patients who worsened by ≥ 10 points from baseline) was lower for cemiplimab versus chemotherapy at cycle 7 in the SCC population (Fig. 3A) and at cycle 8 in the overall population (Fig. 3B). The median time to definitive deterioration of GHS/QoL was longer with cemiplimab than with chemotherapy in both populations (Fig. 4A and B).
      Fig. 3
      Fig. 3Responder analyses, using missing-at-random imputation. GHS/QoL and functioning scales in (A) the SCC population at cycle 7 and (B) the overall population at cycle 8. Symptoms in (C) the SCC population at cycle 7 and (D) the overall population at cycle 8. Cemi, cemiplimab; Chemo, chemotherapy; GHS, global health status; QoL, quality of life; SCC, squamous cell carcinoma.
      Fig. 4
      Fig. 4Time to definitive deterioration. GHS/QoL in (A) the SCC population and (B) the overall population. Physical functioning in (C) the SCC population and (D) the overall population. Pain in (E) the SCC population and (F) the overall population. Fatigue in (G) the SCC population and (H) the overall population. CI, confidence interval; GHS, global health status; HR, hazard ratio; NYR, not yet reached; QoL, quality of life; SCC, squamous cell carcinoma.

      3.3 Physical functioning

      MMRM results for physical functioning showed that at each cycle, LS mean scores were generally maintained or improved with cemiplimab, while those for chemotherapy generally worsened (95% CIs did not include null) in both populations (Fig. 1C and D). In the SCC population, the overall LS mean difference between treatment groups in change from baseline for physical functioning was statistically significant in favour of cemiplimab (8.35; 95% CI: 4.08–12.62; P < 0.0001; Fig. 2A). The treatment difference at cycle 2 favoured cemiplimab (4.09; 95% CI: 0.54–7.64; P = 0.012). In the overall population, treatment differences favoured cemiplimab for overall change from baseline (8.26; 95% CI: 4.29–12.22; P < 0.0001) (Fig. 2B) and at cycle 2 (4.16; 95% CI: 1.08–7.23; P = 0.0041).
      Sensitivity analysis using PMM showed clinically meaningful LS mean differences between treatment arms in favour of cemiplimab across multiple delta values in both populations (Supplementary Fig. 4A and 4B). The proportion of responders was higher and the proportion of non-responders was lower for cemiplimab compared with chemotherapy in the SCC population at cycle 7 (Fig. 3A) and in the overall population at cycle 8 (Fig. 3B). The median time to definitive deterioration of physical functioning was longer with cemiplimab than with chemotherapy in both populations (Fig. 4C and D).

      3.4 Other functioning scales

      Differences between treatment groups in overall LS mean changes from baseline favoured cemiplimab for role functioning, emotional functioning and social functioning in both populations (Figs. 2A and B). Treatment differences for role functioning exceeded the clinically meaningful threshold of 10 points in each population. Treatment differences favoured cemiplimab for all functioning scales in PMM sensitivity analyses, with clinically meaningful differences for role functioning in the SCC population (Supplementary Fig. 4A) and for role functioning, emotional functioning and social functioning in the overall population (Supplementary Fig. 4B). The proportion of responders was higher (Fig. 3A and B) and the median time to definitive deterioration was longer with cemiplimab than with chemotherapy (Supplementary Figs. 5A and 5B) for all functioning scales in both populations.

      3.5 Symptoms

      Treatment differences for fatigue, nausea/vomiting, pain, insomnia, appetite loss and constipation all favoured cemiplimab in both populations (Figs. 2C and D). Differences in appetite loss and pain exceeded the clinically meaningful threshold. In PMM sensitivity analyses, treatment differences favoured cemiplimab for the same symptom scales in the SCC population (Supplementary Fig. 4C) and the same scales plus dyspnoea in the overall population (Supplementary Fig. 4D). The proportion of responders generally was higher for cemiplimab versus chemotherapy for these symptoms in the SCC population through cycle 7 (Fig. 3C) and in the overall population through cycle 8 (Fig. 3D). The median time to definitive deterioration was longer with cemiplimab than with chemotherapy in both populations for pain (Figs. 4E and F), fatigue (Figs. 4G and H) and nausea/vomiting, dyspnoea (overall population only), insomnia, appetite loss and constipation (Supplementary Figs. 5C and 5D).

      4. Discussion

      The primary analysis of Study R2810-ONC-1676 (EMPOWER-Cervical 1/GOG-3016/ENGOT-cx9) demonstrated significant improvement in clinical end-points for cemiplimab versus chemotherapy in patients with recurrent or metastatic cervical cancer, including significant reduction in the risk of death and disease progression in both the SCC and overall populations [
      • Tewari K.S.
      • Monk B.J.
      • Vergote I.
      • Miller A.
      • de Melo A.C.
      • Kim H.S.
      • et al.
      Survival with cemiplimab in recurrent cervical cancer.
      ]. In this analysis, patients showed low-to-moderate symptom burden at baseline, with functioning scores that were lower than normative values in the general population [
      • Hinz A.
      • Singer S.
      • Brahler E.
      European reference values for the quality of life questionnaire EORTC QLQ-C30: results of a German investigation and a summarizing analysis of six European general population normative studies.
      ] and slightly lower than reference values for patients with cancer.
      Cemiplimab provided statistically significant benefit versus chemotherapy in the SCC population for overall changes from baseline in GHS/QoL and physical functioning. Treatment differences also favoured cemiplimab over chemotherapy for changes from baseline in GHS/QoL and physical functioning in the overall population. Early benefits of cemiplimab over chemotherapy were seen at cycle 2 in both populations, but the statistical hierarchy broke with the evaluation of changes from baseline to cycle 2 for GHS/QoL. Separation was seen by cycle 3 or 4, suggesting that comparisons at cycle 2 may have been premature. The proportion of patients who reported a ≥10-point GHS/QoL or physical functioning improvement (a clinically meaningful within-patient change on the EORTC QLQ-C30 scale [
      • King M.T.
      The interpretation of scores from the EORTC quality of life questionnaire QLQ-C30.
      ,
      • Osoba D.
      • Rodrigues G.
      • Myles J.
      • Zee B.
      • Pater J.
      Interpreting the significance of changes in health-related quality-of-life scores.
      ,
      • Cocks K.
      • King M.T.
      • Velikova G.
      • Fayers P.M.
      • Brown J.M.
      Quality, interpretation and presentation of European Organisation for Research and Treatment of Cancer quality of life questionnaire core 30 data in randomised controlled trials.
      ]) was higher, and the proportion with ≥10-point worsening was lower, with cemiplimab compared with chemotherapy. The median time to definitive deterioration (≥10-point sustained worsening) was longer with cemiplimab than with chemotherapy for both GHS/QoL and physical functioning.
      Other functioning scales showed treatment differences favouring cemiplimab for role functioning, emotional functioning and social functioning in both populations (SCC and overall), including a clinically meaningful benefit for role functioning. Symptom scores favoured cemiplimab for fatigue, nausea/vomiting, pain, insomnia, appetite loss and constipation. Time to definitive deterioration also favoured cemiplimab for fatigue, pain, nausea/vomiting, insomnia, appetite loss and constipation. The reported treatment differences for pain and appetite loss exceeded the clinically meaningful threshold. QoL, physical functioning, role functioning and disease-related symptoms such as pain are recommended by regulatory authorities as core outcomes for clinical trials of treatment for cancer [
      European Medicines Agency
      Reflection paper on the regulatory guidance for the use of health-related quality of life (HRQL) measures in the evaluation of medicinal products. Doc. Ref. EMEA/CHMP/EWP/139391/2004.
      ,
      • Food and Drug Administration
      Core patient-reported outcomes in cancer clinical trials: Guidance for industry. Silver Spring.
      ].
      Study R2810-ONC-1676 is the largest phase III randomised, controlled clinical trial of second-line treatment for recurrent or metastatic cervical cancer and the first such study to examine PROs. A single-arm study reported that PROs were stable after 9 weeks of nivolumab treatment in 19 women with cervical cancer [
      • Naumann R.W.
      • Hollebecque A.
      • Meyer T.
      • Devlin M.J.
      • Oaknin A.
      • Kerger J.
      • et al.
      Safety and efficacy of nivolumab monotherapy in recurrent or metastatic cervical, vaginal, or vulvar carcinoma: results from the phase I/II CheckMate 358 trial.
      ] but that study lacked a control arm or long-term monitoring of changes.
      Some limitations should be considered. EORTC QLQ-C30 was designed as a generic cancer instrument and may not capture all pertinent symptoms or components of QoL for patients with cervical cancer. It has nonetheless shown adequate psychometric measurement properties in cervical cancer [
      • Perwitasari D.A.
      • Atthobari J.
      • Dwiprahasto I.
      • Hakimi M.
      • Gelderblom H.
      • Putter H.
      • et al.
      Translation and validation of EORTC QLQ-C30 into Indonesian version for cancer patients in Indonesia.
      ] and is one of the most commonly used PRO instruments in cancer research. Future research should continue to examine the effects of cemiplimab and other new treatments on PROs by also using questionnaires that are specifically designed for use in this patient population. Study treatment was administered open-label but several studies have shown that the potential for bias in PROs may be less prominent in open-label studies than is commonly assumed [
      • Mouillet G.
      • Efficace F.
      • Thiery-Vuillemin A.
      • Charton E.
      • Van Hemelrijck M.
      • Sparano F.
      • et al.
      Investigating the impact of open label design on patient-reported outcome results in prostate cancer randomized controlled trials.
      ,
      • Atkinson T.M.
      • Wagner J.S.
      • Basch E.
      Trustworthiness of patient-reported outcomes in unblinded cancer clinical trials.
      ]. PRO completion rates in this study were high and similar in both treatment arms, suggesting low bias. There was a higher dropout rate in the chemotherapy arm due to differences in progression and survival in favour of cemiplimab. Per the PRO statistical analysis plan, between-group comparisons of PRO data were limited to on-treatment analyses. Continued comparison of PROs after the discontinuation of study treatment would have limited generalisability because it would be confounded by the effects of subsequent treatments on QoL and symptoms. Sensitivity analysis with PMM partly addressed this concern by showing that differences also favoured cemiplimab when data were assumed to be missing-not-at-random. A 10-point threshold was used for clinically meaningful differences, based on prior research in other cancer types [
      • King M.T.
      The interpretation of scores from the EORTC quality of life questionnaire QLQ-C30.
      ,
      • Osoba D.
      • Rodrigues G.
      • Myles J.
      • Zee B.
      • Pater J.
      Interpreting the significance of changes in health-related quality-of-life scores.
      ,
      • Cocks K.
      • King M.T.
      • Velikova G.
      • Fayers P.M.
      • Brown J.M.
      Quality, interpretation and presentation of European Organisation for Research and Treatment of Cancer quality of life questionnaire core 30 data in randomised controlled trials.
      ], but it has not been specifically established for use in patients with cervical cancer. Other analyses that were beyond the scope of this study should be considered for future research of PROs in patients with cervical cancer who receive cemiplimab or other immunotherapy, such as PROs by PD-L1 status and other patient subgroups, or the application of analytical methods such as quality-adjusted time without symptoms or toxicity (Q-TWIST) [
      • Solem C.T.
      • Kwon Y.
      • Shah R.M.
      • Aly A.
      • Botteman M.F.
      Systematic review and benchmarking of quality-adjusted time without symptoms or toxicity (Q-TWiST) in oncology.
      ] or quality-adjusted progression-free survival [
      • Hilpert F.
      • Du Bois A.
      Patient-reported outcomes in ovarian cancer: are they key factors for decision making?.
      ].
      In this international, randomised controlled study, treatment with cemiplimab resulted in a statistically significant benefit versus chemotherapy in the SCC population for GHS/QoL and physical functioning. Consistent benefits over chemotherapy were observed across most functioning and symptom scales, meeting the clinically meaningful threshold for between-group differences in role functioning, appetite loss and pain. The primary analysis showed that cemiplimab significantly improved overall survival, and this analysis showed that PROs further supported the favourable benefit–risk profile of cemiplimab compared with chemotherapy in patients with recurrent or metastatic cervical cancer.

      Funding

      This work was supported by Regeneron Pharmaceuticals, Inc. , and Sanofi.

      Role of the funding source

      The funder participated in study design, data analysis and interpretation, and manuscript writing, and maintained the study database. All authors had full access to the data and had final responsibility for the decision to submit for publication.

      Data sharing

      Qualified researchers may request access to study documents (including the clinical study report, study protocol with any amendments, blank case report form, and statistical analysis plan) that support the methods and findings reported in this manuscript. Individual anonymised participant data will be considered for sharing once the product and indication have been approved by major health authorities (e.g. Food and Drug Administration, European Medicines Agency), if there is legal authority to share the data, and there is not a reasonable likelihood of participant re-identification. Requests should be submitted to https://vivli.org/.

      Conflict of interest statement

      The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: A.O. reports serving on advisory boards for Roche, AstraZeneca, MSD/Merck, PharmaMar, Clovis Oncology, Tesaro, Immunogen, Genmab, Mersana Therapeutic, GSK, SUTRO, AGENUS and Deciphera Pharmaceuticals; and support for travel or accommodation from Roche, AstraZeneca and PharmaMar.
      B.J.M. reports consulting honoraria from Aravive, Asymmetric Therapeutics, Boston Biomedical, ChemoCare, ChemoID, Circulogene, Conjupro Biotherapeutics, Eisai, Geistlich, Genmab/Seattle Genetics, Gynecologic Oncology Group Foundation, Immunogen, Immunomedics, Incyte, Laekna Health Care, Mateon/Oxigene, Merck, Mersana, Myriad, Nucana, Oncomed, Oncoquest, Oncosec, Perthera, Pfizer, Precision Oncology, Puma, Regeneron, Samumed, Takeda, VBL and Vigeo; and consulting or speaker honoraria from AstraZeneca, Clovis, Janssen/Johnson & Johnson, Roche/Genentech and Tesaro/GSK.
      I.V. reports consulting fees from AstraZeneca, Elevar Therapeutics, Genmab, GSK, Immunogen, Merck Sharp & Dohme and Oncoinvent; and contracted research from Genmab and Hoffmann-La Roche.
      A.C.M. has served on advisory boards for Merck Sharp & Dohme, Bristol-Myers Squibb and Libbs; has received support for travel or accommodation from AstraZeneca, Merck Sharp & Dohme, Bristol-Myers Squibb and Roche; and reports institutional research grant support from Clovis Oncology, Bristol-Myers Squibb, Roche, Novartis, Amgen, Merck Sharp & Dohme, Lilly, Pierre Fabre, Sanofi and Pfizer.
      Y.M.K. reports ownership of stock at Johnson & Johnson and Genolution for self and spouse; consulting or advisory role at Merck Sharp & Dohme; and research funding from Regeneron Pharmaceuticals, Inc., and Roche.
      A.S.L., V.S., H.S.K., E.A.G., F.D. and C.-L.C. declare no conflict of interest.
      S.T. reports honoraria from Daiichi Sankyo, Eisai, Bayer, Taiho Pharmaceutical, Merck Sharp & Dohme, Novartis, Chugai Pharma, AstraZeneca, Bristol-Myers Squibb Japan, Ono Pharmaceutical, Nihonkayaku, Pfizer and Lilly Japan; advisory role at Bayer; research funding from Daiichi Sankyo, Sanofi, Eisai, Bayer, Taiho Pharmaceutical, Merck Sharp & Dohme, Novartis, Chugai Pharma, AstraZeneca, Bristol-Myers Squibb, Lilly, Ono Pharmaceutical, PharmaMar and Pfizer/EMD Serono; and travel, accommodation and expenses from Daiichi Sankyo and Novartis.
      J.L., M.M., M.G.F., I.L., J.H. and C.C. are employees and shareholders of Regeneron Pharmaceuticals, Inc.
      C.I. and M.R. report employment at IQVIA and institutional research funding from Regeneron Pharmaceuticals, Inc.
      P.R.L. is an employee of, and may hold shares and stock options in Sanofi.
      K.T. reports honoraria from Tesaro and Clovis Oncology; consulting or advisory roles for Genentech, Tesaro, Clovis and AstraZeneca; serving on a speaker's bureau for Genentech, AstraZeneca, Merck, Tesaro and Clovis; institutional research funding from AbbVie, Genentech, Morphotek, Merck and Regeneron Pharmaceuticals, Inc.; and travel, accommodation and expenses from Genentech.

      Acknowledgements

      The authors thank the patients and their families and caregivers, as well as the primary investigators and site personnel, for participating in the study. Medical writing support was provided by Jonathan Latham of PharmaScribe, LLC, and editorial support was provided by Prime Global. This assistance was funded by Regeneron Pharmaceuticals, Inc., and Sanofi.

      Appendix A. Supplementary data

      The following is the Supplementary data to this article:

      References

        • Sung H.
        • Ferlay J.
        • Siegel R.L.
        • Laversanne M.
        • Soerjomataram I.
        • Jemal A.
        • et al.
        Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.
        CA A Cancer J Clin. 2021; 71: 209-249https://doi.org/10.3322/caac.21660
        • American Society of Clinical Oncology
        Cervical cancer: Statistics.
        2021
        • Corkum M.T.
        • Shaddick H.
        • Jewlal E.
        • Patil N.
        • Leung E.
        • Sugimoto A.
        • et al.
        When Pap testing fails to prevent cervix cancer: a qualitative study of the experience of screened women under 50 with advanced cervix cancer in Canada.
        Cureus. 2019; 11e3950https://doi.org/10.7759/cureus.3950
        • Herzog T.J.
        • Wright J.D.
        The impact of cervical cancer on quality of life--the components and means for management.
        Gynecol Oncol. 2007; 107: 572-577https://doi.org/10.1016/j.ygyno.2007.09.019
        • Chase D.M.
        • Huang H.Q.
        • Wenzel L.
        • Cella D.
        • McQuellon R.
        • Long H.J.
        • et al.
        Quality of life and survival in advanced cervical cancer: a Gynecologic Oncology Group study.
        Gynecol Oncol. 2012; 125: 315-319https://doi.org/10.1016/j.ygyno.2012.01.047
        • Addario B.
        • Geissler J.
        • Horn M.K.
        • Krebs L.U.
        • Maskens D.
        • Oliver K.
        • et al.
        Including the patient voice in the development and implementation of patient-reported outcomes in cancer clinical trials.
        Health Expect. 2020; 23: 41-51https://doi.org/10.1111/hex.12997
        • Bhatnagar V.
        • Hudgens S.
        • Piault-Louis E.
        • Jones L.
        • Beaver J.A.
        • Lyerly H.K.
        • et al.
        Patient-reported outcomes in oncology clinical trials: stakeholder perspectives from the accelerating anticancer agent development and validation workshop 2019.
        Oncol. 2020; 25: 819-821https://doi.org/10.1634/theoncologist.2020-0062
        • Tewari K.S.
        • Monk B.J.
        • Vergote I.
        • Miller A.
        • de Melo A.C.
        • Kim H.S.
        • et al.
        Survival with cemiplimab in recurrent cervical cancer.
        N Engl J Med. 2022; 386: 544-555https://doi.org/10.1056/NEJMoa2112187
        • Fayers P.M.
        • Aaronson N.K.
        • Bjordal K.
        • Groenvold M.
        • Curran D.
        • Bottomley A.
        • et al.
        The EORTC QLQ-C30 scoring manual.
        3rd ed. EORTC Quality of Life Group, Brussels, Belgium2001
        • King M.T.
        The interpretation of scores from the EORTC quality of life questionnaire QLQ-C30.
        Qual Life Res. 1996; 5: 555-567https://doi.org/10.1007/BF00439229
        • Osoba D.
        • Rodrigues G.
        • Myles J.
        • Zee B.
        • Pater J.
        Interpreting the significance of changes in health-related quality-of-life scores.
        J Clin Oncol. 1998; 16: 139-144https://doi.org/10.1200/JCO.1998.16.1.139
        • Cocks K.
        • King M.T.
        • Velikova G.
        • Fayers P.M.
        • Brown J.M.
        Quality, interpretation and presentation of European Organisation for Research and Treatment of Cancer quality of life questionnaire core 30 data in randomised controlled trials.
        Eur J Cancer. 2008; 44: 1793-1798https://doi.org/10.1016/j.ejca.2008.05.008
        • Chung H.C.
        • Ros W.
        • Delord J.P.
        • Perets R.
        • Italiano A.
        • Shapira-Frommer R.
        • et al.
        Efficacy and safety of pembrolizumab in previously treated advanced cervical cancer: results from the phase II KEYNOTE-158 study.
        J Clin Oncol. 2019; 37: 1470-1478https://doi.org/10.1200/JCO.18.01265
        • Hinz A.
        • Singer S.
        • Brahler E.
        European reference values for the quality of life questionnaire EORTC QLQ-C30: results of a German investigation and a summarizing analysis of six European general population normative studies.
        Acta Oncol. 2014; 53: 958-965https://doi.org/10.3109/0284186X.2013.879998
        • Scott N.W.
        • Fayers P.
        • Aaronson N.K.
        • Bottomley A.
        • de Graeff A.
        • Groenvold M.
        • et al.
        EORTC QLQ-C30 reference values.
        2nd ed. EORTC Quality of Life Group, Brussels, Belgium2008
        • European Medicines Agency
        Reflection paper on the regulatory guidance for the use of health-related quality of life (HRQL) measures in the evaluation of medicinal products. Doc. Ref. EMEA/CHMP/EWP/139391/2004.
        Committee for Medicinal Products for Human Use (CHMP), London, UK2005
        • Food and Drug Administration
        Core patient-reported outcomes in cancer clinical trials: Guidance for industry. Silver Spring.
        US Department of Health and Human Services, MD2021
        • Naumann R.W.
        • Hollebecque A.
        • Meyer T.
        • Devlin M.J.
        • Oaknin A.
        • Kerger J.
        • et al.
        Safety and efficacy of nivolumab monotherapy in recurrent or metastatic cervical, vaginal, or vulvar carcinoma: results from the phase I/II CheckMate 358 trial.
        J Clin Oncol. 2019; 37: 2825-2834https://doi.org/10.1200/JCO.19.00739
        • Perwitasari D.A.
        • Atthobari J.
        • Dwiprahasto I.
        • Hakimi M.
        • Gelderblom H.
        • Putter H.
        • et al.
        Translation and validation of EORTC QLQ-C30 into Indonesian version for cancer patients in Indonesia.
        Jpn J Clin Oncol. 2011; 41: 519-529https://doi.org/10.1093/jjco/hyq243
        • Mouillet G.
        • Efficace F.
        • Thiery-Vuillemin A.
        • Charton E.
        • Van Hemelrijck M.
        • Sparano F.
        • et al.
        Investigating the impact of open label design on patient-reported outcome results in prostate cancer randomized controlled trials.
        Cancer Med. 2020; 9: 7363-7374https://doi.org/10.1002/cam4.3335
        • Atkinson T.M.
        • Wagner J.S.
        • Basch E.
        Trustworthiness of patient-reported outcomes in unblinded cancer clinical trials.
        JAMA Oncol. 2017; 3: 738-739https://doi.org/10.1001/jamaoncol.2016.3328
        • Solem C.T.
        • Kwon Y.
        • Shah R.M.
        • Aly A.
        • Botteman M.F.
        Systematic review and benchmarking of quality-adjusted time without symptoms or toxicity (Q-TWiST) in oncology.
        Expert Rev Pharmacoecon Outcomes Res. 2018; 18: 245-253https://doi.org/10.1080/14737167.2018.1434414
        • Hilpert F.
        • Du Bois A.
        Patient-reported outcomes in ovarian cancer: are they key factors for decision making?.
        Expert Rev Anticancer Ther. 2018; 18: 3-7https://doi.org/10.1080/14737140.2018.1516146