CADTH Reimbursement Review

Pembrolizumab (Keytruda)

Sponsor: Merck Canada Inc.

Therapeutic area: Renal cell carcinoma, adjuvant treatment

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Stakeholder Input

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Table 1: Submitted for Review

NOC = Notice of Compliance; RCC = renal cell carcinoma.

Introduction

Renal cell carcinoma (RCC) accounts for more than 80% of all cases of kidney cancer across the globe.^1,2 Histologically, clear cell RCC is the most common form of RCC, accounting for 80% of all cases observed in clinical practice.^1,3 In 2021, kidney and renal pelvis cancers were reported to be the seventh most common cancers in Canada among males (5,200 new cases; 2.8% disease-related deaths) and the 12th most common among females (2,600 new cases; 1.7% disease-related deaths).⁴ Approximately 65% of individuals are typically diagnosed when their tumour is confined to a primary site (local disease), whereas a smaller proportion of patients is diagnosed when the tumour has spread to regional lymph nodes or metastatic sites (16% at regional and 16% at distant stages).⁵ Because of the widespread use of noninvasive abdominal imaging, approximately 50% of kidney tumours are detected incidentally, and many are asymptomatic.^2,3,6 Classic symptoms (flank pain, visible hematuria, and palpable abdominal mass) are usually associated with more advanced disease stages and poorer prognoses.^2,6 Survival rates among patients with RCC largely depend on clinical factors, such as tumour stage, grade, RCC subtype, presence of sarcomatoid features, local extent of the tumour, presence of regional nodal metastasis, and evidence of metastatic disease at presentation.² Scoring systems — such as stage, size, grade, and necrosis (SSIGN) scoring and the University of California Los Angeles Integrated Staging System — classify patients as low, intermediate, or high risk following nephrectomy.^3,7 Estimated rates of 5-year metastasis-free survival among individuals at low, intermediate, and high SSIGN risk are more than 95%, approximately 80%, and less than 40%, respectively.^3,8

In Canada, the current standard of care for nonmetastatic RCC is nephrectomy.⁹ Adjuvant treatment is not recommended in patients with nonmetastatic RCC after nephrectomy, according to the Kidney Cancer Research Network of Canada (KCRNC).¹⁰ Moreover, there were no Health Canada–approved adjuvant treatment options for patients with RCC following nephrectomy at the time of this CADTH review. The current oncologic standard of care for these patients is “observation.”

Pembrolizumab is a high-affinity humanized monoclonal antibody that exerts dual ligand blockade of the programmed cell death receptor 1 (PD-1) pathway, enhances immune system detection of tumours, and facilitates tumour regression. Pembrolizumab is available as a solution for IV infusion in a 100 mg per 4 mL vial and as a 50 mg powder for reconstitution in a single-use vial. Recommended dosing for patients at intermediate-high or high risk of recurrence following nephrectomy or following nephrectomy and resection of metastatic lesions is 200 mg every 3 weeks or 400 mg every 6 weeks until disease recurrence, unacceptable toxicity or up to 1 year (12 months) or 17 doses for 200 mg or 9 doses for 400 mg, whichever is longer, in patients without disease progression.¹¹

Pembrolizumab received Health Canada approval on August 18, 2022, for the adjuvant treatment of adult patients with RCC at intermediate-high or high risk of recurrence following nephrectomy or following nephrectomy and resection of metastatic lesions.

The objective of this review is to perform a systematic review of the beneficial and harmful effects of pembrolizumab for the adjuvant treatment of RCC in patients with intermediate-high or high risk of recurrence following nephrectomy or following nephrectomy and resection of metastatic lesions.

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient groups that responded to CADTH’s call for patient input and from clinical experts consulted by CADTH for the purpose of this review.

Patient Input

One response to CADTH’s call for patient input was received from Kidney Cancer Canada (KCC), which is a national community that provides support and education for patients living with kidney cancer and advocates for their care. The submission was based on 2 online surveys conducted by KCC in 2018 and March 2022, 1 online survey conducted by KCC in collaboration with the International Kidney Cancer Coalition (IKCC) in 2020, and 1 online survey conducted by IKCC in May 2021. Moreover, direct input was collected in March 2022 from 1 American patient with RCC who had experience with pembrolizumab. The 2018 survey supported a previous submission reviewed by CADTH by reporting on the challenges met by patients and caregivers living with kidney cancer. Among the 2,012 respondents to the 2020 international survey, 241 were from Canada; of these, 205 (85%) were patients with kidney cancer, 34 (14%) were caregivers, and the status of the other 2 (0.8%) was undisclosed. A total of 141 patients with RCC responded to the 2021 survey. Of the 106 respondents to the 2022 survey, 65 (61%) were patients or caregivers from Canada.

KCC reported that a large proportion of patients with RCC may eventually experience disease recurrence after nephrectomy, leading to a substantially shortened life expectancy. The patient group input indicated that in the absence of adjuvant therapy options, patients at intermediate to high risk of recurrence experience anxiety and emotional distress from the expectation of recurrence and progression of disease. Nearly half (49%) of survey respondents indicated they would accept adjuvant immunotherapy if it reduced the risk of disease recurrence by 40% to 50%. Approximately 50% of respondents indicated they would accept the risk of side effects associated with steroid use, in the range of 20% to 25%, to manage the side effects of adjuvant immunotherapy.

According to the 1 patient who had experience with pembrolizumab in the adjuvant setting, the side effects of the treatment, including slight occasional rash, slight fatigue, and hyperkalemia, are manageable.

KCC emphasized that there is currently an unmet need for an effective adjuvant therapy for kidney cancer that reduces the risk of disease recurrence and improves patient outcomes by, for example, reducing the number of patients who develop metastatic disease and the costs associated with RCC care.

A copy of the patient input from KCC is presented in the Stakeholder section of this review.

Clinician Input

Input From Clinical Experts Consulted by CADTH

The clinical experts consulted by CADTH emphasized that there is an unmet need among patients with a higher risk of recurrent disease after surgery for kidney cancer. Currently, there is no approved adjuvant treatment in this setting. The experts reported that pembrolizumab would be offered as monotherapy in the adjuvant setting after resection of kidney cancer in patients at intermediate-high or high risk of recurrence. In the opinion of the clinical experts consulted, patients with clear cell carcinoma with M1 (distant metastasis) resected metastases would benefit most from adjuvant treatment, followed by patients with the pathological T3 (pT3) to pathological T4 (pT4) tumours (those at high risk of recurrence), and patients with T2 grade 3 to 4 tumours (those at intermediate risk of recurrence). The experts identified patients with autoimmune diseases requiring steroids as those who should not receive adjuvant pembrolizumab. The clinical experts noted that overall survival (OS) and disease-free survival (DFS) are important outcome measures for the assessment of patient’s response to treatment. Discontinuation of treatment was recommended by the clinical experts in the case of disease recurrence or intolerable treatment toxicities. The experts reported that treatment administration and monitoring should be undertaken by a medical oncologist in an outpatient or community cancer setting.

Clinician Group Input

Two clinician groups provided input for this review: the Kidney Cancer Research Network of Canada and the Ontario Health Genitourinary Cancer Drug Advisory Committee. Two clinicians affiliated with the Kidney Cancer Research Network of Canada and 1 clinician on behalf of Ontario Health Genitourinary Cancer Drug Advisory Committee contributed to this submission. The clinician groups agreed that there is an unmet need for adjuvant therapy to lower the risk of disease recurrence in patients with localized RCC following nephrectomy in Canada. The clinician groups indicated that, if funded, pembrolizumab would be the first adjuvant therapy option for patients with RCC in Canada.

A copy of the clinician input from the Kidney Cancer Research Network of Canada and Ontario Health Genitourinary Cancer Drug Advisory Committee is presented in the Stakeholder section of this review.

Drug Program Input

Drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may affect their ability to implement a recommendation. The drug plans anticipated that adjuvant pembrolizumab might cause a shift in subsequent therapy lines that are reimbursed in the Canadian setting. The drug plans anticipated changes to the dosing modality in practice, with changes toward weight-based dosing schedule for pembrolizumab. Implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 4.

Clinical Evidence

Pivotal and Protocol-Selected Study

Description of Study

The KEYNOTE-564 trial is an ongoing multicentre, randomized, double-blind, phase III study with a primary objective to compare the efficacy and safety of pembrolizumab versus placebo as an adjuvant treatment for adult patients with RCC postnephrectomy or post nephrectomy and resection of metastatic lesions. The trial was conducted at 212 sites in 21 countries, including Canada. The study enrolled patients aged 18 years and older with a histologically confirmed diagnosis of RCC with a clear cell component with or without sarcomatoid features. The study included patients at intermediate-high or high risk of recurrence, based on pathological tumour, node, metastasis (TNM) staging, Fuhrman grade, and the presence of sarcomatoid features, and patients following metastatic disease who had undergone complete resection of primary and metastatic lesions. Patients were also required to be tumour-free, have an Eastern Cooperative Oncology Group Performance Status (ECOG PS) of 0 or 1, and have no prior systemic treatment for RCC.¹² The primary outcome investigated in the KEYNOTE-564 trial was DFS, assessed by the investigator. The key secondary outcome was OS, and other secondary outcomes included disease recurrence–specific survival (DRSS), event-free survival (EFS) assessed by blinded independent radiology review, safety, and health-related quality of life (HRQoL).¹²

Patients were randomized in 1:1 ratio to receive either pembrolizumab (200 mg IV infusion every 3 weeks) or placebo (saline solution IV every 3 weeks) for up to a maximum of 17 infusions or (approximately) 1 year, until confirmation of recurrence, treatment discontinuation, or study termination. Randomization was based on metastasis status variable (M0 [no distant metastasis] versus M1 NED [distant metastasis with no evidence of disease] defined as primary kidney tumour plus solid, isolated, soft tissue metastases that were completely resected at nephrectomy [synchronous] or no more than 1 year after nephrectomy [metachronous]), and within the M0 group on following stratification factors: ECOG PS (0 or 1) and US participant (yes or no). By the first interim analysis (December 14, 2020), 1,406 patients were screened and 994 were randomized into the trial (496 to pembrolizumab and 488 to placebo arms). One additional analysis (efficacy update report) was implemented after 6 additional months of follow-up, with a cut-off date of June 14, 2021.

The median age of patients enrolled in the KEYNOTE-564 study was 60 years, and the majority of participants were White men. Most patients had tumours without sarcomatoid features and with intermediate-high risk for recurrence. Baseline characteristics were equally balanced in the 2 study arms. More patients discontinued treatment in the pembrolizumab arm (38.9%) than in the placebo arm (26.2%), primarily due to adverse events (AEs). More patients in the placebo arm (22.5%) received subsequent systemic anti-cancer treatment compared with the pembrolizumab arm (15.3%).¹²

Efficacy Results

The key efficacy findings of the KEYNOTE-564 study are summarized in Table 2.

Overall Survival

At the first interim analysis data cut-off (December 14, 2020), the median follow-up durations were 24 months (range = 2.5 to 41.5 months) and 23.8 months (range = 3.5 to 41.4 months) for patients in pembrolizumab and placebo arms, respectively. Median OS was not reached in either treatment arm. A hazard ratio (HR) of 0.54 (95% CI, 0.30 to 0.96; P = 0.0164) was estimated for the comparison between pembrolizumab and placebo. Additional 6-month follow-up data from the efficacy update report analysis (June 14, 2021, data cut-off) showed that median OS was not reached in either groups, with an observed HR of 0.52 (95% CI, 0.31 to 0.86; P = 0.005).

DFS Assessed by Investigator

Similarly, median DFS was not reached in either treatment group at the time of the first interim analysis (December 14, 2020). The HR obtained for pembrolizumab versus placebo was 0.68 (95% CI, 0.53 to 0.87; P = 0.001). In the efficacy update report analysis, with a data cut-off date of June 14, 2021, the HR was 0.63 (95% CI, 0.50 to 0.80; P < 0.0001). Median DFS was not reached in either group at the time of the efficacy update report analysis.

According to the pre-specified subgroup analysis, HRs for DFS in the metastatic staging groups was 0.74 (95% CI, 0.57 to 0.96) for the M0 group and 0.29 (95% CI, 0.12 to 0.69) for the M1 NED group. Similar findings were observed in the efficacy update report analysis, which had 6 additional months of follow-up; for the M0 subgroup, the HR was 0.68 (95% CI, 0.53 to 0.88), and for the M1 NED subgroup, the HR was 0.28 (95% CI, 0.12 to 0.66). The efficacy update report results for the post hoc subgroup analysis of recurrence risk showed the following estimates: HR = 0.68 (95% CI, 0.52 to 0.89) for intermediate-high risk, HR = 0.60 (95% CI, 0.33 to 1.10) for high risk, and HR = 0.28 (95% CI, 0.12 to 0.66) for M1 NED risk.

Health-Related Quality of Life

HRQoL assessments included the overall least squares mean difference estimated for the pembrolizumab and placebo arms. Among patients completing the health-related quality of life measures, patients in both arms appeared to experience a slight deterioration in HRQoL and symptom worsening assessed at week 52. The overall least squares mean difference in Functional Assessment of Cancer Therapy Kidney Symptom Index–Disease-Related Symptoms (FKSI-DRS) score was –0.67 (95% CI, –1.23 to –0.12). The least squares mean difference in the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) questionnaire was –2.57 (95% CI, –5.22 to 0.08) for global health status/quality of life (QoL) scale and –0.91 (95% CI, –2.79 to 0.97) for the physical functioning scale.

Harms Results

The proportion of patients with at least 1 treatment-emergent adverse event appeared higher in the pembrolizumab arm (96.3%) than in the placebo group (91.1%). Serious adverse events (SAEs) were reported in 20.5% of individuals who received pembrolizumab treatment compared with 11.3% of individuals receiving placebo. There were more AEs leading to drug discontinuations (pembrolizumab versus placebo: 20.7% versus 2.0%) and more treatment interruptions in the pembrolizumab group (25.8% versus 14.9%). Overall, 2 deaths were reported in the pembrolizumab arm (0.4%) and 1 death was reported in the placebo arm (0.2%).

Notable harms were higher in the pembrolizumab group than in the placebo group, including hyperthyroidism (21.1% versus 6.9%) hypothyroidism (11.9% versus 0.2%), pneumonitis (2.3% versus 1%), adrenal insufficiency (2% versus 0.2%), type 1 diabetes mellitus (1.8% versus 0%), colitis (1.6% versus 0.2%), severe skin reactions (1.6% versus 0.4%), infusion reactions (1.4% versus 1%), thyroiditis (1.2% versus 0.2%), and hepatitis (1% versus 0%).

Table 2: Summary of Key Results From the Pivotal and Protocol-Selected Study (KEYNOTE-564)

AE = adverse event; APaT = all participants as treated; CFB = change from baseline; CI = confidence interval; DFS = disease-free survival; EORTC QLQ-C30 = European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FAS = full analysis set; FKSI-DRS = Functional Assessment of Cancer Therapy Kidney Symptom Index–Disease-Related Symptoms; HR = hazard ratio; ITT = intention to treat; LSM = least squares mean; NR = not reached; OS = overall survival; SAE = serious adverse events; SD = standard deviation.

Notes: For PROs at baseline and week 52, n is the number of participants in each treatment group with no missing assessments at the specific time point; for change from baseline, n is the number of participants in the analysis population in each treatment group.

Data cut-off: December 14, 2020.

^aFrom product-limit (Kaplan-Meier) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (yes vs. no) within M0 group by investigator.

^cOne-sided P value based on log-rank test stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (yes vs. no) within M0 group by investigator.

^dBased on a cLDA model with the PRO scores as the response variable with covariates for treatment by study visit interaction, stratification factors metastasis status (M0 vs. M1 NED), and within M0 group further stratified by ECOG PS (0 vs. 1) and US participant (yes vs. no) as covariates.

Source: Clinical Study Report.¹²

Critical Appraisal

The KEYNOTE-564 trial is an ongoing multicentre, randomized, placebo-controlled, double-blind study. The randomization scheme implemented in the trial minimized the risk of bias due to unknown confounders. Because of the placebo-controlled design, unblinding might potentially have occurred due to higher frequencies of immune-related AEs in the pembrolizumab compared with the placebo arm. Baseline and demographic characteristics were balanced in the 2 study arms, suggesting successful randomization. Concomitant medications permitted in the trial, as well as subsequent anti-cancer therapies administered, were considered appropriate by the clinical experts consulted by CADTH and were reflective of treatments used in Canadian practice.

OS, DFS, and HRQoL, which were investigated in the KEYNOTE-564 trial, were considered clinically meaningful outcomes by the clinical experts and reflective of outcomes assessed in clinical practice. Other surrogate end points, such as DRSS and EFS, were considered to be of lower clinical relevance, according to the clinical experts.

The primary outcome (DFS) was assessed by local investigators, and blinded independent central review (BICR) assessments were introduced to evaluate the robustness of the DFS findings. Findings of DFS by BICR were consistent with the primary analysis, suggesting a low possibility of evaluation bias. Multiplicity adjustments were implemented adequately for the analysis of DFS and OS, and sensitivity analyses were also pre-specified and conducted for DFS. The findings from the sensitivity analyses were consistent with the primary intention-to-treat (ITT) analyses. Median DFS and OS were not reached at the time of the interim analyses, suggesting data immaturity. More patients in the placebo arm received post-treatment anti-cancer therapies compared with the pembrolizumab arm, which might have produced biased estimates of OS (favouring the placebo group). Of note, surgery in patients with RCC is performed with curative intent, and rates of 5-year disease-specific survival are high in patients at intermediate (about 80%) and high risk of recurrence (from 40% to 55%) postnephrectomy. Hence, longer follow-up is needed to observe the effects of adjuvant pembrolizumab on survival outcomes. Findings from the analysis of secondary and exploratory outcomes (EFS, DRSS, HRQoL), as well as defined subgroups, were considered exploratory, as no multiplicity adjustments were performed. The magnitude of effect of pembrolizumab on the HRQoL of patients in the adjuvant setting is uncertain because of the lack of formal hypothesis testing, possible violation of missing data assumptions in the model applied, and high attrition rates.

There were several interim analyses pre-specified in the protocol before the first interim analysis (December 14, 2020), which was used as the base for this CADTH report. Another interim analysis (i.e., efficacy update report) with 6 months of additional follow-up data was added between the first and the second interim analysis to respond to potential requests from regulatory agencies. The final OS analysis will take place after approximately 200 deaths are observed between the pembrolizumab and placebo groups. Adjustments were made to account for alpha spending in the interim analyses.

The clinical experts consulted by CADTH for this review reported that baseline characteristics and findings from KEYNOTE-564 can be generalized to adults with RCC living in Canada who are at intermediate-high or high risk of recurrence following nephrectomy or following nephrectomy and resection of metastatic lesions. The administered dosage of pembrolizumab was 200 mg every 3 weeks for up to 17 doses, which is aligned with the approved Health Canada indication. The clinical experts noted that dosing of 400 mg every 6 weeks for up to 9 doses is more commonly applied in real clinical practice. The appropriateness of placebo as the comparator was confirmed by the clinical experts because there are no Health Canada–approved adjuvant treatment options available in Canada. According to the clinical experts, patients recruited in the pivotal trial had more frequent disease assessments and follow-up procedures compared with what would be applied in patients in real-world practice.

Conclusions

One sponsor-submitted, multicentre, randomized, double-blind, phase III trial (KEYNOTE-564) comparing adjuvant pembrolizumab with placebo in patients with RCC was included in this CADTH systematic review.

Overall, pembrolizumab improved DFS outcome, compared with placebo, as an adjuvant treatment for patients with RCC who are at intermediate-high or high risk of recurrence after nephrectomy or following nephrectomy and resection of metastatic lesions. However, the effects of adjuvant pembrolizumab relative to placebo on OS could not be determined because of the immature survival data, uncertain influence of subsequent treatments, and uncertainty about the correlation between DFS and OS in the adjuvant treatment of RCC. Likewise, limitations with the HRQoL analyses in the single randomized controlled trial precluded the ability to draw conclusions about the effects of pembrolizumab on this outcome. The safety profile of pembrolizumab was similar to that observed in other trials of this drug, including effects on the thyroid and adrenal glands. The clinical experts indicated that baseline characteristics and findings from the KEYNOTE-564 trial could be generalized to patients with RCC in the adjuvant setting in Canada.

Introduction

Disease Background

RCC is the most common form of kidney cancer, accounting for more than 80% of all cases in the world.^1,2 Histologically, RCCs are further classified into different subtypes: clear cell, papillary, chromophobe, clear cell papillary, collecting duct, medullary, and unclassified.^1,3 Almost all types of RCC can have or develop sarcomatoid features, but they are reported most often in chromophobe histological subtypes.¹³ Clear cell RCC is the most common form, accounting for approximately 80% of all RCC cases in clinical practice.^3,5 The estimated proportions of individuals diagnosed at the local (confined to primary site), regional (spread to regional lymph nodes), and distant (metastasized cancer) stages are 65%, 16%, and 16%, respectively.⁵

With the widespread use of noninvasive abdominal imaging, almost half of all kidney tumours are detected incidentally, many of which are asymptomatic.^2,3,6 The classic symptoms of flank pain, visible hematuria, and palpable abdominal mass are associated with more advanced disease stages and poorer prognosis.^2,6 Common diagnostic methods for identifying and characterizing tumours and assessing disease progression include CT scans, MRI, X-rays, and bone scans.^3,5 Known risk factors for RCC include smoking, hypertension, obesity, medications (over-the-counter pain killers, phenacetin-containing compounds, and diuretics), family history of RCC, genetic conditions (von Hippel-Lindau disease), and hereditary papillary RCC.^5,14

The Canadian epidemiological estimates from 2021 showed that kidney and renal pelvis cancers were the seventh most common cancers in males (accounting for 5,200 new cases and 2.8% of disease-related deaths) and the 12th most common in females (2,600 new cases and 1.7% of disease-related deaths). The predicted 5-year age-standardized survival was 73% for both sexes. The predicted net survival was higher for patients within the age group of 15 to 44 years (92%) compared with patients aged 85 years and older (33%).⁴ The 5-year survival rate depends on clinical, anatomic, and histological factors, such as tumour stage, grade, RCC subtype, presence of sarcomatoid features, local extent of tumour, presence of regional nodal metastasis, and evidence of metastatic disease at presentation.² The 5-year relative survival rates for patients with localized, regional, or distant kidney or renal pelvis cancer was 92.7%, 71%, and 13.9%, respectively.⁵

Different scoring systems that assess the probability of survival in patients postnephrectomy, based on histological and clinical features, have been proposed.³ SSIGN scoring is based on TNM stage and different pathologic features (size, nuclear grade, and presence of necrosis). The estimated 5-year metastasis-free survival among individuals at low, intermediate, and high risk is estimated to be more than 95%, approximately 80%, and less than 40%, respectively.^3,8 Another scoring model, the University of California Los Angeles Integrated Staging System, incorporates TNM staging, ECOG PS, and Fuhrman pathological grading to classify individuals with localized and metastatic disease into low-, intermediate-, and high-risk subgroups following nephrectomy.^3,7 The 5-year disease-specific survival for individuals with localized disease is estimated to be 91.1% for patients at low risk, 80.4% for patients at intermediate risk, and 54.7% for patients at high risk.³ For patients with metastatic disease, survival probabilities are worse, at 32%, 19.5% and 0% in the low-, intermediate-, and high-risk subgroups, respectively).³

Standards of Therapy

In the Canadian clinical context, the current standard of care for nonmetastatic RCC is nephrectomy.⁹ The most recent consensus statement from the KCRNC does not recommend adjuvant therapy in patients with nonmetastatic RCC following nephrectomy, but encourages patient participation in clinical trials whenever possible.¹⁰ At the time of this CADTH review, there were no Health Canada–approved adjuvant treatment options for patients with RCC following nephrectomy. The current oncologic standard of care for these patients is observation.

According to the clinical experts consulted by CADTH, important treatment goals for patients on adjuvant treatment include prolonging survival and reducing the risk of recurrence with no negative impact on quality of life.

Drug

Pembrolizumab is a high-affinity humanized monoclonal antibody that blocks the PD-1 pathway (Table 3).

Health Canada has issued market authorization for pembrolizumab for the treatment of various types of tumours.¹¹

Pembrolizumab received Health Canada approval for the following indication: adjuvant treatment of adult patients with RCC at intermediate-high or high risk of recurrence following nephrectomy, or following nephrectomy and resection of metastatic lesions. The CADTH reimbursement request aligns with the approved Health Canada indication.

Table 3: Key Characteristics of Pembrolizumab

PD-1 = programmed cell death; PD-L1 = programmed cell death ligand-1; PD-L2 = programmed cell death ligand-2; RCC = renal cell carcinoma.

^aHealth Canada proposed indication.

Source: Health Canada product monograph.¹¹

Stakeholder Perspectives

Patient Group Input

This section was prepared by CADTH staff based on the input provided by patient groups. The full patient group input is included in the Stakeholder Input section of this review.

One response to CADTH’s call for patient input was received for this review: a submission from KCC. KCC is a national community that provides support and education to individuals living with kidney cancer and advocates for their care.

The information in the submission was based on 4 online surveys: 2 conducted by KCC in 2018 and in March 2022, 1 conducted in collaboration with the IKCC in 2020, and 1 conducted by IKCC in May 2021. Additionally, direct input was collected in March 2022 from 1 American patient with RCC who had experience with pembrolizumab. The 2018 survey supported a previous submission reviewed by CADTH by reporting on the challenges faced by patients and caregivers living with kidney cancer. Among the 2,012 respondents of the 2020 international survey, 241 were from Canada; of these, 205 (85%) were patients with kidney cancer, 34 (14%) were caregivers, and the status of the other 2 (0.8%) was undisclosed. A total of 141 patients with RCC responded to the 2021 survey. Among the 106 respondents to the 2022 survey, 65 (61%) patients and caregivers were from Canada.

KCC described the experience of patients with kidney cancer as they navigate the health care system. Notably, KCC stated that a large proportion of patients with RCC may eventually experience disease recurrence after nephrectomy, leading to a substantially shortened life expectancy. KCC also noted that there are currently no available treatment options funded in Canada for adjuvant therapy to reduce the risk of disease recurrence, and in the absence of adjuvant therapy options, patients at intermediate to high risk of recurrence experience anxiety and emotional distress related to concerns about recurrence and progression of disease following nephrectomy. Nearly half (49%) of survey respondents indicated that they would accept adjuvant immunotherapy postsurgery if the therapy reduced the risk of disease recurrence by 40% to 50%. Approximately 50% of respondents indicated that they would accept the risk of side effects associated with steroid use to manage the side effects of adjuvant immunotherapy if that level of risk is in the range of 20% to 25%. Approximately 51% of respondents equated DFS with a “longer period of time that I remain cancer free on surveillance scans.” More than half (56%) of respondents equated OS with “longer survival – living a longer life, even if my kidney cancer eventually does return.”

According to the 1 patient who had experience with pembrolizumab for the indication under review, the side effects of treatment, which included slight occasional rash, slight fatigue, and hyperkalemia, were manageable. KCC emphasized that there is currently an unmet need for an effective adjuvant therapy for kidney cancer to reduce the risk of disease recurrence and improve patient outcomes by, for example, reducing the number of patients who develop metastatic disease and the costs associated with RCC care.

A copy of the patient input from KCC is presented in the Stakeholder Input section of this review.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise in the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy). The following input was provided by 2 clinical specialists with expertise in the diagnosis and management of RCC.

Unmet Needs

In Canada, there are currently no approved adjuvant treatments following surgery for kidney cancer. Clinical experts highlighted that there is an unmet need among patients at a higher risk of recurrent disease after nephrectomy. Important treatment goals would include lengthening survival of these patients and reducing the risk of recurrence, without negatively influencing patients’ quality of life.

Place in Therapy

The clinical experts consulted by CADTH reported that pembrolizumab would be considered monotherapy in the adjuvant setting in patients who have undergone resection of kidney cancer who are at intermediate-high or high risk of recurrence. The experts believed that adjuvant pembrolizumab would cause a shift in treatment paradigm for a couple of reasons. First, clinicians are not currently offering adjuvant treatment in this setting. Medical oncologists would be required to offer balanced and informed discussions about adjuvant pembrolizumab. Second, downstream treatment options may be affected in cases in which a patient develops metastatic disease. For instance, if a patient experiences relapse while on adjuvant treatment, the clinical experts suggested using tyrosine kinase inhibitors (TKIs)-based agents (e.g., sunitinib, cabozantinib, pazopanib, and axitinib) instead of ipilimumab plus nivolumab or axitinib plus pembrolizumab. In the case of disease recurrence after at least 6 months of adjuvant treatment, the experts suggested re-treatment with a PD-1 combination (e.g., ipilimumab plus nivolumab or axitinib plus pembrolizumab).

Patient Population

In the opinion of the clinical experts consulted, patients with M1 resected metastases would benefit the most from adjuvant pembrolizumab, followed by pT3 to pT4 patient population (who are at high risk of recurrence) and T2, grade 3 to 4 patient population (who are at intermediate risk of recurrence). There is no sufficient evidence to support adjuvant treatment with pembrolizumab in patients with kidney cancer histologies other than clear cell. Moreover, patients at low risk of recurrence would not require adjuvant treatment. One of the clinical experts noted that patients with autoimmune disease who require steroids should not receive adjuvant pembrolizumab, whereas patients with mild autoimmune diseases (not requiring steroids > 10 mg prednisone equivalent) would require a thorough risk-benefit discussion before considering adjuvant pembrolizumab treatment. The clinical experts noted that identification of patients eligible for adjuvant treatment would be done by urologists following surgery, and referral to a medical oncologist would be required. Both experts indicated that misdiagnosis or underdiagnosis is very unlikely because of tangible pathological reporting and the fact that these patients undergo definitive surgery. The experts highlighted the lack of diagnostic biomarkers that would identify the patients most likely to respond to adjuvant treatment.

Assessing Response to Treatment

In terms of patients’ response to treatment, the clinical experts reported that both OS and DFS are important outcomes. When assessing treatment response, the clinical experts consulted by CADTH reported that the frequency of treatment assessments should be aligned with the pivotal clinical trial (i.e., CT scans of the chest, abdomen, and pelvis every 12 weeks during the first year, every 16 weeks in years 2 to 4, and every 24 weeks after 5 years). However, the experts also noted that, in clinical practice, tumour assessments might be conducted less frequently.

Discontinuing Treatment

According to the clinical experts, treatment with pembrolizumab in adjuvant setting should be discontinued in the case of disease recurrence or intolerable treatment toxicities (e.g., myocarditis, encephalitis, severe colitis requiring steroids). One clinical expert noted that discontinuation of treatment would not be required if patients experienced thyroid or hormonal dysfunction, mild rash, or mild colitis.

Prescribing Conditions

The clinical experts consulted by CADTH indicated that treatment administration and the monitoring of patients on adjuvant pembrolizumab should be undertaken by a qualified medical oncologist experienced in immuno-oncologic treatment and capable of identifying and handling immune-related adverse events (irAEs). The experts advised that adjuvant pembrolizumab be administered in an outpatient cancer centre or community cancer centre, and that an on-call service be available to manage irAEs that occur after hours.

Additional Considerations

Both clinical experts highlighted the unmet need for patients at highest risk for recurrence (M1 resected NED, T3 to T4 high grade 3 to 4/4). One expert noted that the previously conducted clinical trials in the adjuvant setting demonstrated limited efficacy of a TKI-based agent, sunitinib, that is currently in use only in the US.

Clinician Group Input

This section was prepared by CADTH staff based on the input provided by clinician groups. The full clinician group input is included in the Stakeholder Input section of this review.

Two clinician groups provided input for this review: the KCRNC and the Ontario Health Genitourinary Cancer Drug Advisory Committee. Two clinicians affiliated with the KCRNC and 1 clinician on behalf of Ontario Health Genitourinary Cancer Drug Advisory Committee contributed to this submission. The clinician groups agreed that there is an unmet need for adjuvant therapy to lower the risk of disease recurrence in patients with localized RCC following nephrectomy in Canada. The clinician groups indicated that, if funded, pembrolizumab would be the first adjuvant therapy option for patients with RCC in Canada.

A copy of the clinician input from the KCRNC and the Ontario Health Genitourinary Cancer Drug Advisory Committee is presented in the Stakeholder Input section of this review.

Drug Program Input

Drug programs provide input on each drug being reviewed through CADTH’s reimbursement review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 4.

Table 4: Summary of Drug Plan Input and Clinical Expert Response

AE = adverse event; BIA = budget impact analysis; ECOG PS = Eastern Cooperative Oncology Group Performance Status; M0 = no distant metastasis; N0 = no regional lymph node metastasis; PAG = Provincial Advisory Group; PD-1 = programmed cell death; pERC = CADTH pan-Canadian Oncology Drug Review Expert Review Committee; pT3 = pathological T3; RCC = renal cell carcinoma; TKI = tyrosine kinase inhibitor.

Clinical Evidence

The clinical evidence included in the review of pembrolizumab is presented in 3 sections. The first section, the systematic review, includes pivotal studies provided in the sponsor’s submission to CADTH and Health Canada and studies that were selected according to an a priori protocol. The second section includes indirect evidence from the sponsor (if submitted) and indirect evidence selected from the literature that met the selection criteria specified in the review. For this submission, the sponsor did not provide indirect evidence. The third section includes sponsor-submitted long-term extension studies and additional relevant studies that were considered to address important gaps in the evidence included in the systematic review.

Systematic Review (Pivotal and Protocol-Selected Studies)

Objectives

To perform a systematic review of the beneficial and harmful effects of pembrolizumab (200 mg, IV, once every 3 weeks; or 400 mg, IV, once every 6 weeks) for the adjuvant treatment of RCC in patients at intermediate-high or high risk of recurrence following nephrectomy, or following nephrectomy and resection of metastatic lesions.

Methods

Studies selected for inclusion in the systematic review included pivotal studies provided in the sponsor’s submission to CADTH and Health Canada and studies meeting the selection criteria presented in Table 5. Outcomes included in the CADTH review protocol reflect outcomes considered to be important to patients, clinicians, and drug plans.

Of note, the systematic review protocol presented in this section was established before the granting of a Notice of Compliance from Health Canada.

Table 5: Inclusion Criteria for the Systematic Review

AE = adverse event; DFS = disease-free survival; EFS = event-free survival; HRQoL = health-related quality of life; M0 = no distant metastases; M1 = distant metastasis; M1 NED = M1 with no evidence of disease; OS = overall survival; RCC = renal cell carcinoma; RCT = randomized controlled trial; SAE = serious adverse event; SSIGN = stage, size, grade, and necrosis scoring; UISS = University of California Los Angeles Integrated Staging System; WDAE = withdrawal due to adverse event.

^aThese outcomes were identified as being of particular importance to patients in the input received by CADTH from patient groups.

The literature search was performed by an information specialist using a peer-reviewed search strategy.

The literature search for clinical studies was performed by an information specialist using a peer-reviewed search strategy, according to the PRESS Peer Review of Electronic Search Strategies checklist.¹⁵

Published literature was identified by searching the following bibliographic databases: MEDLINE All (1946—) via Ovid and Embase (1974—) via Ovid. All Ovid searches were run simultaneously as a multi-file search. Duplicates were removed using Ovid deduplication for multi-file searches, followed by manual deduplication in Endnote. The search strategy comprised both controlled vocabulary, such as the National Library of Medicine’s MeSH (Medical Subject Headings), and keywords. The main search concepts were Keytruda (pembrolizumab) and RCC. Clinical trials registries were searched: the US National Institutes of Health’s clinicaltrials.gov, WHO’s International Clinical Trials Registry Platform search portal, Health Canada’s Clinical Trials Database, and the European Union Clinical Trials Register.

No filters were applied to limit retrieval by study type. Retrieval was not limited by publication date or by language. Conference abstracts were excluded from the search results. Refer to Appendix 1 for the detailed search strategies.

The initial search was completed on April 1, 2022. Regular alerts updated the search until the meeting of the CADTH Canadian Drug Expert Committee (CDEC) on August 10, 2022.

Grey literature (literature that is not commercially published) was identified by searching relevant websites from the Grey Matters: A Practical Tool For Searching Health-Related Grey Literature reference.¹⁶ Included in this search were the websites of regulatory agencies (US FDA and European Medicines Agency). Google was used to search for additional internet-based materials. Refer to Appendix 1 for more information on the grey literature search strategy.

The manufacturer of the drug was contacted for information regarding unpublished studies.

Two CADTH clinical reviewers independently selected studies for inclusion in the review based on titles and abstracts, according to the predetermined protocol. Full-text articles of all citations considered potentially relevant by at least 1 reviewer were acquired. Reviewers independently made the final selection of studies to be included in the review, and differences were resolved through discussion.

Findings From the Literature

One study was identified for inclusion in the systematic review (Figure 1). The included study is summarized in Table 6. A list of excluded studies is presented in Appendix 2.

Figure 1: Flow Diagram for Inclusion and Exclusion of Studies

Table 6: Details of Included Studies

AE = adverse event; BICR = blinded independent central review; DFS = disease-free survival; DRSS = disease recurrence–specific survival; EFS = event-free survival; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ-5D-5L = 5-Level EQ-5D; FKSI-DRS = Functional Assessment of Cancer Therapy Kidney Symptom Index–Disease-Related Symptoms; IA1 = first interim analysis; IV = intravenously; M0 = patients with no distant metastases; M1 NED = patients with a primary kidney tumour plus solid, isolated, soft tissue metastases that were completely resected at nephrectomy (synchronous) or ≤ 1 year after nephrectomy (metachronous); MRI = MRI; OS = overall survival; PD-1 = programmed cell; PD-L1 = programmed cell death ligand-1; PD-L2 = programmed cell death ligand-2; PRO = patient-reported outcome; RCC = renal cell carcinoma; SAE = serious adverse event; TMN = tumour, node, and metastasis staging; VAS = visual analogue scale.

^aFor participants who discontinue treatment intervention and who will not enter the efficacy follow-up phase, the first survival follow-up contact will be scheduled 12 weeks after discontinuation visit and/or the safety follow-up visit (whichever is last). For participants who completed assessments in the efficacy follow-up phase, the first survival follow-up contact will be scheduled 12 weeks after the last efficacy follow-up visit had been performed.

Source: Clinical Study Report.¹²

Description of Studies

KEYNOTE-564 is an ongoing multicentre, randomized, double-blind, phase III study with a primary objective to compare the efficacy and safety of pembrolizumab versus placebo (saline solution) as an adjuvant treatment for adult patients with RCC postnephrectomy or postnephrectomy and resection of metastatic lesions.

The trial was conducted at 212 global sites in 21 countries (including Canada). The key characteristics of the study design are summarized in Table 6 and Figure 2.

Stratified randomization using central Interactive Voice Response Systems and Interactive Web Response Systems was implemented at participating trial sites. Patients were stratified by metastasis status (M0 versus M1 NED), which was predefined. Moreover, in the M0 group, 2 additional stratification factors were pre-specified: ECOG PS (0 versus 1) and US participant (yes versus no).

The first patient was randomized on June 30, 2017, and the last patient was enrolled September 20, 2019. The cut-off date for first interim analysis was December 14, 2020, and database lock date was January 26, 2021. One additional analysis (efficacy update report) was implemented after 6 additional months of follow-up, with a cut-off date of June 14, 2021.

Figure 2: KEYNOTE-564 Study Design

mo = month; NED = no evidence of disease; Q3W = every 3 weeks.

** Safety follow-up: 30 days after last dose; efficacy follow-up: every 12, 16, or 24 weeks; survival follow-up: every 12 weeks. Survival follow-up status is assessed approximately every 12 weeks until death, withdrawal of consent, or the end of the study, whichever occurs first. For participants who discontinue treatment intervention and who will not enter the efficacy follow-up phase, the first survival follow-up contact will be scheduled 12 weeks after the discontinuation visit and/or the safety follow-up visit (whichever is last). For participants who completed assessments in the efficacy follow-up phase, the first survival follow-up contact will be scheduled 12 weeks after the last efficacy follow-up visit had been performed.

Source: Clinical Study Report.¹²

Populations

Inclusion and Exclusion Criteria

Eligible patients included individuals aged 18 years and older with histologically confirmed clear cell RCC, with or without sarcomatoid features, and with intermediate-high or high risk of recurrence following nephrectomy or following nephrectomy and resection of metastatic lesions. Recurrence risk categories were defined by pathological TNM staging, Fuhrman grade, and presence of sarcomatoid features (Table 6). Patients could not have received prior systemic treatment for RCC. Patients with prior major surgery other than nephrectomy and/or resection of pre-existing metastases for M1 NED disease, and those with residual thrombus postnephrectomy in the renal vein or vena cava were excluded from the trial. Moreover, patients who had received prior systemic treatment or radiotherapy for RCC or who had pre-existing brain or bone metastasis were not eligible.

Baseline Characteristics

As of the first interim analysis data cut-off (December 14, 2020), the median age of patients randomized in the KEYNOTE-564 study was 60 years (range for pembrolizumab group = 27 to 81 years; range for placebo group = 25 to 84 years), more males than females were enrolled (70.0% and 72% in pembrolizumab and placebo arms, respectively), and the majority of patients were White (75.0% and 75.7% in pembrolizumab and placebo arms, respectively). Most patients had tumours with an absence of sarcomatoid features and were at intermediate-high risk of recurrence. Overall, baseline characteristics were balanced in the 2 study arms. Table 7 presents the baseline summary characteristics of the ITT population.

Regarding prior treatment, 1 participant had received prior anti-cancer drug therapy (pazopanib hydrochloride), and 95% of participants in both the pembrolizumab and placebo arms had at least 1 concomitant medical condition. The most frequently reported medical conditions in participants receiving pembrolizumab were hypertension (53%) and type 2 diabetes mellitus (12%), and in participants receiving placebo were hypertension (50%), gastroesophageal reflux disease (12%), back pain (12%), benign prostatic hyperplasia (12%), and hyperlipidemia (11%).

Table 7: Summary of Patient Baseline Characteristics (ITT Population) in the KEYNOTE-564 Trial

CPS = combined positive score; ECOG PS = Eastern Cooperative Oncology Group Performance Statis; ITT = intention to treat; M0 = patients with no distant metastases; M1 NED = patients with a primary kidney tumour plus solid, isolated, soft tissue metastases that were completely resected at nephrectomy (synchronous) or ≤ 1 year after nephrectomy (metachronous); PD-L1 = programmed cell death ligand-1.

Notes: Participants in the M0-intermediate-high risk category have pT2 (grade 4 or sarcomatoid), N0, M0 disease; or pT3 (any grade), N0, M0 disease. Participants in the M0-high-risk category have pT4 (any grade), N0, M0 disease; or pT any (any grade), N1 or greater, M0 disease. Participants in the M1 NED category present not only with the primary kidney tumour, but also with solid, isolated, soft tissue metastases that were completely resected at the time of nephrectomy (synchronous) or 1 year from nephrectomy (metachronous). Participants in the M0-other category have T2 (grade ≤ 3) N0, M0 disease; or T1, N0, M0 disease.

Data cut-off: December 14, 2020.

Source: Clinical Study Report.¹²

Interventions

Patients were randomized to receive 1 of the following treatments:

• pembrolizumab (200 mg as a 30-minute IV infusion every 3 weeks)

• placebo (saline solution IV every 3 weeks).

Study treatments were administered to a maximum of 17 infusions (or up to approximately 1 year) on an outpatient basis or until confirmation of disease recurrence or the criteria for treatment discontinuation are met.

Concomitant Medication

Patients were prohibited from receiving the following therapies during the trial: antineoplastic systemic chemotherapy or biologic therapy, immunotherapy or chemotherapy not specified in the trial protocol, investigational drugs other than pembrolizumab, radiation therapy, live vaccines in the 30 days before the first dose of the study treatment (e.g., measles, mumps, rubella, varicella-zoster, yellow fever, rabies, Bacillus Calmette-Guérin, typhoid vaccine), systemic glucocorticoids for any purpose other than to modulate symptoms from an AE suspected to have an immunologic etiology. Details of treatment with other, permitted concomitant medications are presented in the Table 14.

Treatment Discontinuation Criteria

Discontinuation of the study treatment was considered for patients who received 17 cycles of study treatment, experienced progression or recurrence of the malignancy under study, experienced a new malignancy requiring active treatment or any other intercurrent illness that prevents administration of the study treatment. Patients were also discontinued from study treatment in case of the development of unacceptable toxicity, development of recurrent grade 2 pneumonitis, positive serum pregnancy test, patients’ request, investigator’s decision, or study termination due to a withdrawal of consent. Discontinuation from the study treatment was considered permanent and patients were not allowed to restart study treatment after their discontinuation.

Outcomes

The primary objective was to assess whether pembrolizumab prolongs DFS, assessed by the investigator, compared with placebo. The key secondary objective was to assess whether pembrolizumab prolongs OS, compared with placebo. Other secondary objectives included a comparison of safety and tolerability, DRSS, EFS by blinded independent radiology review, DFS and OS according to programmed death ligand-1 (PD-L1) expression, EORTC QLQ-C30 (global health status/QoL and physical functioning scales) and FKSI-DRS scores in the pembrolizumab and placebo arms. Exploratory objectives included pharmacokinetic parameters, the presence of antidrug antibodies, biomarker analyses, EORTC QLQ-C30 symptom subscale scores, 5-Level EQ-5D (EQ-5D-5L) visual analogue scale (VAS) scores, and the proportion of patients experiencing deterioration, stability, improvement, and stability and improvement, as indicated by EORTC QLQ-C30 and FKSI-DRS scores.

A list of efficacy end points identified in the CADTH review protocol that were assessed in the clinical trials included in this review is provided in Table 8, and detailed summaries of the end points follow. A detailed discussion and critical appraisal of outcome measures are provided in Appendix 4.

Table 8: Summary of Outcomes of Interest in the KEYNOTE-564 Trial Identified in the CADTH Review Protocol

AE = adverse event; BICR = blinded independent central review; DFS = disease-free survival; DRSS = disease recurrence–specific survival; EFS = event-free survival; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; EQ-5D-5L = 5-Level EQ-5D; FKSI-DRS = Functional Assessment of Cancer Therapy Kidney Symptom Index–Disease-Related Symptoms; HRQoL = health-related quality of life; M0 = patients with no distant metastases; M1 = distant metastasis; M1 NED = M1 with no evidence of disease; OS = overall survival; RCC = renal cell carcinoma; SAE = serious adverse event; VAS = visual analogue scale.

Efficacy Measurement for Primary and Secondary Outcomes

Tumour assessments (consisting of CT or MRI scans of the brain, chest, abdomen, and pelvis and a bone scan) were performed within 28 days before randomization. For each patient, the same imaging modality and image-acquisition protocol was used consistently across all time points.

During the treatment period, patients who discontinued the study treatment for a reason other than disease recurrence were considered to be on study and continued with scheduled assessments. During treatment, imaging assessments of the chest, abdomen, and pelvis were performed every 12 weeks from randomization. On-treatment bone and brain scans were performed only as clinically indicated during the postrandomization period.

Safety follow-up visits were conducted 30 days (± 7 days) after the last dose of the study treatment or before the initiation of new anti-cancer treatment, whichever occurred first.

All individuals who completed 17 cycles of the study treatment or discontinued treatment for reasons other than disease recurrence entered the efficacy follow-up and were monitored for disease status on the following schedule: every 12 weeks in year 1, every 16 weeks in years 2 to 4, and every 24 weeks in year 5 and beyond. Disease imaging continued until disease recurrence, pregnancy, the start of a new anti-cancer treatment, withdrawal of consent, death, or the end of the trial, whichever occurred first.

For the assessment of DFS, radiographic evidence of disease recurrence was required and biopsy or cytology confirmation was strongly encouraged. The date of disease recurrence was the date the first radiographic image was obtained, regardless of the timing of additional confirmatory procedures (e.g., repeat imaging, biopsy, and/or cytology). Disease progression was assessed in participants considered to have baseline disease by BICR and was defined as the unequivocal progression of baseline disease or the appearance of new lesions, assessed by BICR.

Patients who completed all efficacy assessments and/or patients who did not have further efficacy assessments (including patients who discontinued treatment and did not enter efficacy follow-up phase) entered the survival follow-up phase. Assessment of survival occurred approximately every 12 weeks until death, withdrawal of consent, or the end of the trial, whichever occurred first.

Health-Related Quality of Life

HRQoL measures were assessed using the generic cancer HRQoL instrument (EORTC QLQ-C30), the RCC-specific HRQoL instrument (FKSI-DRS), and the generic HRQoL instrument (EQ 5D-5L). The HRQoL instruments were completed electronically by study participants at various time points: during treatment (cycles 1, 5, 9, 13, and 17), at treatment discontinuation, at 30-day follow-up, and annually during the post-treatment period until disease recurrence or the start of a new anti-cancer treatment. A detailed discussion and critical appraisal of these outcomes are available in Appendix 4.

EORTC QLQ-C30

EORTC QLQ-C30 is a self-reported cancer-specific, multidimensional questionnaire that assesses HRQoL associated with treatment in patients with cancer.¹⁸ This 30-item questionnaire incorporates 5 multi-item functional scales (physical, role, cognitive, emotional, and social), 3 multi-item symptom scales (fatigue, pain, nausea and vomiting), 6 single-item symptom scales (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial impact) and a 2-item quality-of-life scale (global QoL).¹⁹ Higher scores on the functional scale, the symptom scale and items, and the global health status/QoL scale indicate a higher level of functioning, a higher level of symptomatology, and a higher level of quality of life, respectively.¹⁹ The validity, reliability, responsiveness, and minimally important differences (MIDs) of the EORTC QLQ-C30 questionnaire have not been documented in patients with RCC in the adjuvant setting (Appendix 4).

FKSI-DRS

FKSI-DRS is a kidney-cancer-specific, self-reported instrument that evaluates disease-related symptoms. The 9-item questionnaire assesses symptoms of kidney cancer deemed to be important to patients with and clinicians treating advanced kidney cancer (lack of energy, pain, weight loss, bone pain, fatigue, shortness of breath, cough, fever, blood in urine).²⁰ Total scores for FKSI-DRS range from 0 (severely symptomatic) to 36 (asymptomatic).²¹ Validity, reliability, and responsiveness to change has been demonstrated in patients with advanced RCC, but not in the adjuvant setting. The MID estimated using different anchors ranged from 0.62 to 3 points for patients with advanced RCC.

EQ-5D-5L VAS

The EQ-5D-5L is a generic, utility-based measure of HRQoL comprising a descriptive system and the VAS. Only the VAS was assessed in the trial. The EQ VAS is a vertical VAS with values between 100 (best imaginable health) and 0 (worst imaginable health) on which patients provide a global assessment of their health.²² The validity, reliability, and responsiveness of the EQ-5D-5L questionnaire and its VAS have not been evaluated in patients with RCC in the adjuvant setting.

HRQoL thresholds defined in the KEYNOTE-564 study were as follows:

• EORTC QLQ-C30 functional and the global health status/QoL score: a change of 10 points or more was considered to be clinically meaningful²³

• FKSI-DRS score: a change of 3 points or more was considered to be clinically meaningful).^20,24

Statistical Analysis

Sample Size and Power Calculation

Sample size calculations were based on the primary end point (DFS) and the main secondary end point (OS). The study plan was to randomize 950 patients in a 1:1 ratio to pembrolizumab or placebo, but the power calculations by the sponsor were performed on 990 participants, which aligned with the actual number of randomized participants.

The study was designed to achieve 95% power at the 1-sided alpha level of 2.5% to detect a statistically significant difference in DFS between pembrolizumab and placebo (HR = 0.67). A total of 332 DFS events were expected to occur, with 1 interim analysis reaching about 80% of the targeted events.

The power to detect a HR of 0.67 and 0.635 at an overall alpha level of 2.5% in OS between the pembrolizumab and placebo arms was 79% and 88%, respectively. This was conditional upon the rejection of the null hypothesis for DFS, targeted 200 OS events and 3 interim analyses reaching about 47%, 66%, and 86% of the target events.

In the calculation of power for the DFS and OS analyses, the sponsor assumed the following: DFS following a Poisson mixture cure rate model with assumed cure rate of 0.3, based on historical data²⁵; a median DFS of 45 months for individuals not cured in the placebo arm; OS following an exponential distribution and a median of 145 months for the placebo arm; 27-month enrolment with monthly accrual of 20 participants, 30 participants, and 1 participant during the first 5 months, from month 6 to month 21, and for the last month, respectively; and a yearly dropout rate of 2% for DFS and 1% for OS.

Multiplicity

Multiplicity testing adjustments included an overall type I error control at 2.5% (1-sided) for the primary end point (DFS) and key secondary end point (OS). The trial incorporated a fixed testing sequence: DFS at alpha level of 2.5% (1-sided) was tested first, and the alpha was passed to OS if the hypothesis test for DFS is declared successful. A group sequential approach (Maurer and Bretz) was implemented to allocate the alpha between the interim and final analyses.

For DFS and OS, the information fraction at each analysis was based on 332 and 200 final planned events, respectively. Lan-DeMets O’Brien-Fleming alpha spending function was used to set the efficacy bounds. Nonbinding futility spending was implemented by controlling the probability of crossing the futility bound under the null hypothesis (total of 1-alpha = 97.5%); a Hwang-Shih-DeCani spending function, with γ = –6 used to set testing boundaries at each analysis. All the bounds were adjusted in line with the actual number of events accrued at each analysis.

Multiplicity adjustments were not made for other secondary or exploratory analyses.

Planned Analysis

There were 3 planned interim analyses and 1 planned final analysis in the KEYNOTE-564 trial (Table 9). The interim efficacy analyses were conducted by an external unblinded statistician and scientific programmer who had no other responsibilities in the study and were reviewed by an external data monitoring committee. Periodic safety monitoring was conducted by the committee, the frequency of which was defined in the committee charter (about every 6 months in the first year and every 12 months after the first year).

Table 9: Summary of Interim and Final Efficacy Analyses

IA1 = first interim analysis; IA2 = second interim analysis; IA3 = third interim analysis; DFS = disease-free survival; OS = overall survival.

Source: Clinical Study Report.^12,26

Of note, additional amendments to the supplemental statistical plan were implemented to incorporate potential requests from regulatory agencies. Specifically, an efficacy update report was added between 2 planned interim analyses (first and second interim analysis). The efficacy update report incorporated 6 additional months of follow-up from the first interim analysis (approximately 66 OS events), with June 14, 2021, as the data cut-off date. Median follow-up duration was 29.9 months (range = 2.5 to 47.5 months) for individuals in pembrolizumab group and 29.2 months (range = 3.5 to 47.4 months) for individuals in the placebo group. All the amendments were made before the data cut-off date, and the results were reviewed by an external data monitoring committee.

The sponsor indicated that the alpha levels would be spent on the efficacy update based on the observed number of events at the efficacy update report analysis, in line with the alpha spending function initially specified in the protocol, and that the alpha spending and interim analysis-final analysis boundary properties at subsequent interim and final analyses would be updated accordingly. Similar to the initial power calculations, the power to detect a HR of 0.67 and 0.635 at an overall alpha level of 2.5% in OS between the pembrolizumab and placebo arms was 79% and 88%, respectively. The amendment reported the following conditions for this power calculation: rejection of the null hypothesis for DFS; 200 OS events targeted; and 3 interim analyses and a possible efficacy update report analysis reaching about 26%, 33%, 66%, and 86% of the target events.

Primary Outcome: DFS

Median DFS for the pembrolizumab and placebo treatment arms were estimated and plotted using the nonparametric Kaplan-Meier (KM) method (Table 10). The primary hypothesis (i.e., treatment difference in DFS) was evaluated by comparing DFS using a stratified log-rank test. The HRs were estimated using a stratified Cox regression model with Efron’s method of tie handling, with a single-treatment covariate. The stratification factors implemented for both the stratified log-rank test and the stratified Cox model included stratification factors used in randomization (metastasis status [M0 versus M1 NED] and, in the M0 group, ECOG PS [0 versus 1] and US participant [yes versus no]). The analysis of DFS was based on the ITT population.

Subgroup Analyses

Pre-specified subgroups analysis was performed for the primary end point (DFS).

Subgroup analyses did not account for multiplicity. This CADTH review identified 3 subgroups of interest in the protocol, 1 of which was included in the pre-specified subgroup analysis (patients with different metastatic status [M0 or M1 NED] at baseline). Moreover, the efficacy update report analysis presented findings of an additional subgroup analysis of interest for this CADTH review (DFS, according to recurrence risk groups), which was not pre-specified in the analysis plan for the trial.

Sensitivity Analyses

A sensitivity analysis was performed to account for the different time-to-event distribution in individuals enrolled with baseline disease, assessed by BICR, compared to those who were disease-free at baseline. The stratified Cox model and stratified log-rank test was implemented with baseline disease status assessed by BICR as an additional stratum. In particular, 6 strata were used (M0 + ECOG PS 0 + US participant versus M0 + ECOG PS 0 + non-US participant versus M0 + ECOG PS 1 + US participant versus M0 + ECOG PS 1 + non-US participant versus M1 NED versus M1 by BICR). Another sensitivity analysis included the assessment of DFS by BICR and the censoring of individuals with baseline disease, assessed by BICR at the randomization date. The robustness of the DFS end point was assessed in a third pre-specified sensitivity analysis with a different set of censoring rules. This sensitivity analysis took into account new anti-cancer treatment or missed disease assessments. Specifically, the sensitivity analysis followed the same principles as the primary analysis, except in the case of events that occurred after 2 consecutive missed disease assessments or after new anti-cancer therapy, in which case censoring was done at the last disease assessment before the missed assessments and the initiation of new anti-cancer therapy. Censoring rules for primary and sensitivity analysis are described in Table 11.

Secondary Outcome: OS

The analytical approach for the main secondary outcome (OS) was similar to the 1 adopted for the primary outcome of the study. Median OS in both treatment arms was estimated and plotted using the nonparametric KM method. A stratified log-rank test was used to assess the treatment difference in survival. HRs were estimated using a stratified Cox regression model with Efron’s method of tie handling, with a single-treatment covariate. The stratification factors implemented for both the stratified log-rank test and the stratified Cox model included stratification factors used in randomization (metastasis status [M0 versus M1 NED] and, in the M0 group, ECOG PS [0 versus 1], and US participant [yes versus no]). The analysis of OS was based on the ITT population.

Other Secondary Outcomes

Details of the statistical approaches used to assess DRSS and EFS are outlined in the Table 10. Briefly, a nonparametric cumulative incidence estimator was used to estimate DRSS curves in the pembrolizumab and placebo treatment groups. The analyses included DRSS1, for which only local recurrence was considered an event, and DRSS2, for which only local disease recurrence with visceral lesion or occurrence of distant kidney cancer metastasis(es) with visceral lesion were considered as events. Censoring rules for DRSS1 and DRSS1 are described in Table 10.

For EFS, the nonparametric KM method was used to estimate the curve in each treatment arm. No formal hypothesis testing was performed, and nominal P values from the stratified log-rank test were provided. The stratified Cox regression model with Efron’s method of tie handling, with a single-treatment covariate, was used for estimation of HRs. The stratification factors implemented for the stratified Cox model included stratification factors used in randomization, with baseline disease status assessed by BICR as an additional stratum.

HRQoL measures included the EORTC QLQ-C30 (global health status/QoL and physical functioning scales) and FKSI-DRS. These were assessed as secondary outcomes through a constrained longitudinal data analysis (cLDA) model, with the scores as the response variable, and the treatment-by-time interaction and stratification factors used for randomization as covariates. Model-based treatment differences in terms of the least squares mean change from baseline and least squares mean values by treatment arm at the baseline and postbaseline time points (52 weeks) were reported. The full analysis set was used to assess patient-reported outcomes (PROs), defined as all randomized patients with at least 1 HRQoL assessment available for the specific end point and who have received at least 1 dose of the study intervention.

Exploratory Outcomes

Exploratory outcomes included mean change from baseline on the EORTC QLQ-C30 symptom subscales (nausea and vomiting, diarrhea symptoms) and EQ-5D-5L VAS scores, using the methodology described for other secondary outcomes.

Furthermore, an assessment of improvement and improvement plus stability rates in patient-reported end points for the 2 treatment groups was assessed (EORTC QLQ-C30 global health status/QoL and physical functioning scales and FKSI-DRS), in line with the definitions presented in the Appendix 3. The stratified Miettinen and Nurminen method was implemented to compute the difference in the overall improvement (or improvement plus stability) rate between the pembrolizumab and placebo groups and its corresponding 95% CI, with strata weighting by sample size and stratification factors used for randomization. Moreover, the exact binomial method of Clopper and Pearson was adopted to calculate point estimates and the corresponding 95% CI by treatment group.

Safety Outcomes

Safety analyses were considered secondary outcomes in the KEYNOTE-564 trial and followed a tiered approach (refer to Table 10 for description). Tier 1 safety end points (subject to inferential testing for statistical significance) were not evaluated in the study, as no enhanced safety signals were expected. Tier 2 parameters were assessed with descriptive statistics and point estimates with 95% CIs for between-group comparisons, whereas tier 3 safety parameters were only evaluated using descriptive statistics.

Table 10: Statistical Analysis of Efficacy End Points in the KEYNOTE-564 Study

AE = adverse events; BICR = blinded independent central review; CI = confidence interval; cLDA = constrained longitudinal data analysis; DFS = disease-free survival; DRSS = disease recurrence–specific survival; ECG = electrocardiogram; ECOG PS = Eastern Cooperative Oncology Group Performance Status; EFS = event-free survival; HR = hazard ratio; HRQoL = health-related quality of life; KM = Kaplan-Meier; LS = least squares; M0 = patients with no distant metastases; M1 NED = patients with a primary kidney tumour plus solid, isolated, soft tissue metastases that were completely resected at nephrectomy (synchronous) or ≤ 1 year after nephrectomy (metachronous); OS = overall survival; PRO = patient-reported outcome; SAE = serious adverse event; SOC = system organ class; vs. = versus.

Note: Tier 1 safety end points were subject to inferential testing for statistical significance with P values and 95% CI provided for between-group comparisons. Because of the known safety signals of pembrolizumab from previous studies, as well as the lack of expectation of an enhanced safety signal in the adjuvant setting (i.e., participants do not have a tumour burden and their existing safety profile might be less intensive), the sponsor did not analyze tier 1 safety end points.

Table 11: Censoring Rules for Derivation of DFS

DFS = disease-free survival.

Source: Clinical Study Report.¹²

Analysis Populations

The efficacy analysis set (ITT population) was used for the primary and key secondary efficacy analyses, and consisted of all randomized patients, regardless of the treatment received.

The safety analysis set (all participants as treated population) consisted of randomized patients who had received at least 1 dose of any study drug, and analysis was based on the as-treated principle.

The HRQoL analysis set (PRO full analysis set population) consisted of all patients who had at least 1 PRO assessment available and received at least 1 dose of the study treatment. Participants were analyzed in the treatment group to which they were randomized.

Protocol Amendments

The original protocol (v1.0) was issued on February 24, 2017. There were 4 protocol amendments made by the first interim analysis data cut-off date of December 14, 2020 (Table 29).

• Amendment 1 (November 2, 2017): At study inception, the inclusion of metachronous and synchronous M1 NED patients was intended; however, during protocol finalization, metachronous was accidentally removed. This amendment corrected the mistake and added minor adjustments to improve protocol clarity, address real-time feedback from investigators, and adjust language inconsistencies.

• Amendment 2 (September 4, 2019): The second interim analysis was retooled as the first interim analysis, where the trigger is 80% DFS events accrued. This would represent roughly a minimum follow-up of 12 months after enrolment is finished.

• Amendment 3 (May 11, 2020): The trigger for the first interim analysis timing was changed; the total number of targeted events for the final analysis of DFS by investigator review (second interim analysis) was changed; a secondary end point (EFS assessed by blinded independent radiology review) was added.

• Amendment 4 (October 13, 2020): The censoring rules were updated, per the most recent oncology standard for an adjuvant study, to follow the ITT rule and remove the pharmacokinetic and antidrug antibodies sample collection.

All protocol amendments were submitted to the appropriate health authorities and institutional review boards and/or independent ethics committees for information and approval, in accordance with local requirements and/or national regulations.

An additional amendment to the supplemental statistical plan was implemented to incorporate additional months of follow-up and potential agency requests. This amendment specified that the alpha would be spent on the efficacy update, based on the observed number of events at the analysis, per the alpha spending function specified in the initial protocol, whereas the alpha spending and interim analysis-final analysis boundary properties at subsequent interim and final analyses would be updated accordingly.

Results

Patient Disposition

Patient disposition in the KEYNOTE-564 study is presented in Table 12. Screening evaluation was conducted to determine whether participants met the entry requirements (inclusion and exclusion criteria) for the trial. In total, 1,406 patients were screened and 994 were randomized in a 1:1 ratio to receive either pembrolizumab or placebo. A total of 412 patients were screened out of the study, and the reported reason for nonrandomization of these participants was screening failure. Specifically, the most frequently reported reasons for screening failure involved a failure to meet the study inclusion criteria, and included the presence of baseline disease (n = 156; 37.9%), failure to meet criteria for intermediate-high risk, high risk, or M1 NED RCC (n = 66; 16.0%), withdrawal of patient consent (n = 65; 15.8%), absence of prior nephrectomy and/or metastasectomy (n = 42; 10.2%), absence of prior partial or radical nephrectomy or metastasectomy with negative surgical margins (n = 25; 6.1%), inadequate organ function (n = 26; 6.3%), and lack of histologically confirmed RCC diagnosis with clear cell component (n = 21; 5.1%).

A total of 496 patients were randomized to receive adjuvant pembrolizumab and 498 were randomized to receive placebo. Of the 994 randomized patients, 8 (1.6%) in the pembrolizumab arm and 2 (0.4%) in placebo arm did not receive their assigned treatment. No reasons were provided for failure to take the assigned medication.

At the time of the data cut-off date of December 14, 2020, 190 (38.9%) patients had discontinued pembrolizumab and 130 (26.2%) had discontinued placebo. The main reasons for discontinuation of pembrolizumab were AEs (n = 104; 21.3%), disease relapse (n = 51; 10.5%), and withdrawal by patient (n = 21; 4.3%), and for discontinuation of placebo were disease relapse (n = 101; 20.4%), AEs (n = 11; 2.2%), and withdrawal by patient (n = 10; 2.0%).

There were no patients receiving pembrolizumab treatment at the time of the first IA, while 1 patient was reported to be still ongoing placebo treatment. The sponsor clarified that the 1 patient for whom study treatment was reported as ongoing in the placebo arm had received the last dose of study treatment but had not completed the study treatment discontinuation visit at the time of the first interim analysis.

Table 12: Patient Disposition in the KEYNOTE-564 Trial (ITT Population)

AE = adverse event; APaT = all participants as treated; ITT = intention to treat.

Notes: If the overall number of participants is displayed in the first row of a section (status in trial and status of study medication in the trial), then that is used as the denominator for the percentage calculation; otherwise, the number of participants in the population is used as the denominator for the percentage calculation.

Database cut-off date: December 14, 2020

Source: Clinical Study Report.¹²

Major Protocol Deviations

Major protocol deviations considered to be clinically important (i.e., with the potential to compromise critical data analyses pertaining to primary efficacy and/or safety end points or participants’ safety) occurred in 2 participants receiving pembrolizumab and 3 participants receiving placebo. The major protocol deviations deemed clinically important were:

• Inclusion and exclusion criteria (4 patients: 2 in the pembrolizumab arm; 2 in placebo arm): 1 participant in the pembrolizumab group and 1 participant in placebo group were randomized with pre-existing brain or bone metastatic lesion(s); 1 participant in the pembrolizumab group and 1 participant in placebo group were randomized with residual thrombus postnephrectomy in the renal vein or vena cava.

• Safety reporting (1 patient in placebo arm): 1 participant receiving placebo had a reportable safety event and/or follow-up safety event information that was not reported per the timelines outlined in the protocol.

Exposure to Study Treatments

At the first data cut-off (December 14, 2020), the median duration on treatment observed was 11.1 months in both the pembrolizumab and placebo arms (range = 0.0 to 14.3 and 0.0 to 15.4, respectively). Treatment exposure in the KEYNOTE-564 study is presented in the Table 13. The median follow-up duration at the first interim analysis for patients in the pembrolizumab arm was 24 months (range = 2.5 to 41.5 months) and for patients in the placebo arm was 23.8 months (range = 3.5 to 41.4 months).

Concomitant Medications

Overall, 94.7% of patients in the pembrolizumab arm and 93.8% in the placebo arm received at least 1 concomitant medication during the study.

The most frequently administered concomitant medications in the pembrolizumab arm were drugs in the Anatomic Therapeutic Chemical pharmacologic subclasses of analgesics (53.9%), drugs that act on the renin-angiotensin system (41.4%), drugs for acid-related disorders (33.8%), and systemic antibacterial drugs (36.7%), and in the placebo arm were analgesics (50.8%), drugs that act on the renin-angiotensin system (35.9%), and lipid-modifying drugs (31.7%). Table 14 outlines the list of antineoplastic drugs reported in any frequency and other concomitant medications reported in at least 15% of patients in either arm.

Anti-Cancer Medications Received by Patients During the Follow-Up Phase

The proportion of patients receiving subsequent anti-cancer treatment was higher in the placebo arm (22.5%) compared to the pembrolizumab arm (15.3%). Subsequent anti-cancer therapies included drug therapy, radiation, and surgery. Anti–vascular endothelial growth factor treatment (n = 56 [11.3%] in pembrolizumab and n = 76 [15.3%] in the placebo arm) and anti-PD-1/PD-L1 checkpoint inhibitor therapies (n = 14 [2.8%] in pembrolizumab and n = 46 [9.2%] in the placebo arm) were the most commonly reported drug therapies. An overview of subsequent anti-cancer therapies received by patients in both study arms is presented in Table 15.

Table 13: Exposure to Study Treatments (APaT Population; IA1)

APaT = all participants as treated; IA1 = first interim analysis; SD = standard deviation.

Note: Database cut-off date: December 14, 2020.

^aDuration of treatment (months) = (date of last dose of study drug – date of first dose of study drug + 1) / 30.4375.

Source: Clinical Study Report.¹²

Table 14: Concomitant Medications (APaT Population, IA1)

APaT = all participants as treated; IA1 = first interim analysis.

Notes: Every participant is counted a single time for each applicable specific concomitant medication. A participant with multiple concomitant medications within a medication category is counted a single time for that category.

Database cut-off date: December 14, 2020

Source: Clinical Study Report.¹²

Table 15: Subsequent Anti-Cancer Therapy (ITT Population, IA1)

IA1 = first interim analysis; ITT = intention to treat; PD-1 = programmed cell death; PD-L1 = programmed cell death ligand-1; RCC = renal cell carcinoma; VEGF = vascular endothelial growth factor; VEGFR = vascular endothelial growth factor receptor.

Notes: The total number of patients in the subcategories under any subsequent anti-cancer drug therapy can exceed the total number of patients with any subsequent anti-cancer drug therapy because a patient can receive multiple types of subsequent anti-cancer drug therapy.

Database cut-off date: December 14, 2020.

^aIncluded pembrolizumab, avelumab, nivolumab, atezolizumab, and durvalumab.

^bIncluded axitinib, pazopanib, sunitinib, sorafenib, cabozantinib, lenvatinib, tivozanib, and bevacizumab.

^cIncluded all anti-cancer drugs other than anti-PD1 and anti-PDL1 therapies and VEGF and VEGFR targeted therapies.

Source: Clinical Study Report.¹²

Efficacy

Only those efficacy outcomes and analyses of subgroups identified in the review protocol are reported here (refer to Appendix 3 for detailed efficacy data). Findings presented in this review were obtained from the first interim analysis (December 14, 2020). Results from the efficacy update report analysis with additional 6 months of follow-up are presented for the primary (DFS) and main secondary (OS) outcomes.

Overall Survival

At the first interim analysis, median OS was not reached. A total of 51 OS events occurred (18 in the pembrolizumab group and 33 in the placebo group). A HR of 0.54 (95% CI, 0.30 to 0.96; P = 0.0164) was estimated for the comparison between pembrolizumab and placebo. Table 16 presents a summary of efficacy findings, and Figure 3 presents the KM plot for OS.

Additional 6-month follow-up data from the efficacy update report analysis showed similar findings. There was a total of 66 events as of June 14, 2021, data cut-off (23 in the pembrolizumab group and 43 in the placebo group). Median OS was not reached, and a HR of 0.52 was reported (95% CI, 0.31 to 0.86; P = 0.005) (Table 30, Figure 8).

Table 16: Overall Survival at IA1 — Efficacy Analysis Set (ITT Population)

CI = confidence interval; HR = hazard ratio; IA1 = first interim analysis; ITT = intention to treat; NR = not reached; OS = overall survival; Q1 = first quartile; Q3 = third quartile.

Database cut-off date: December 14, 2020.

^aFrom product-limit (KM) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by metastasis status (M0 vs. M1 NED assessed by investigator) and ECOG PS (0 vs. 1), US participant (yes vs. no) within M0 group by investigator.

^cOne-sided P value based on log-rank test stratified by metastasis status (M0 vs. M1 NED assessed by investigator) and ECOG PS (0 vs. 1), US participant (yes vs. no) within M0 group by investigator.

Source: Clinical Study Report.¹²

Figure 3: Kaplan-Meier Plot of OS at IA1 — Full Analysis Set

+ Censored = censored observations; IA1 = first interim analysis; KM = Kaplan-Meier; OS = overall survival.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

DFS, Assessed by Investigator

At the preplanned first interim analysis, 260 DFS events had occurred (109 [22.0%] in pembrolizumab arm and 151 [30.3%] in placebo arm). Median DFS was not reached in either group. The HR between pembrolizumab and placebo was 0.68 (95% CI, 053 to 0.87; P = 0.001) (Table 17, Figure 4)

At the efficacy update report analysis (data cut-off date of June 14, 2021), 114, and 169 DFS events had occurred in the pembrolizumab and placebo arms, respectively. Median DFS was not reached, and a HR of 0.63 (95% CI, 0.50 to 0.80; P < 0.0001) was reported (Table 31, Figure 9).

Table 17: Disease-Free Survival at IA1 — Efficacy Analysis Set (ITT Population)

CI = confidence interval; DFS = disease-free survival; HR = hazard ratio; IA1 = first interim analysis; ITT = intention to treat; NR = not reached; Q1 = first quartile; Q3 = third quartile.

Note: Database cut-off date: December 14, 2020.

^aFrom product-limit (KM) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by metastasis status (M0 vs. M1 NED assessed by investigator) and ECOG PS (0 vs. 1) and US participant (yes vs. no) within M0 group by investigator.

^cOne-sided P value based on log-rank test stratified by metastasis status (M0 vs. M1 NED assessed by investigator) and ECOG PS (0 vs. 1), US participant (yes vs. no) within M0 group by investigator.

Source: Clinical Study Report.¹²

Figure 4: Kaplan-Meier Plot of DFS at IA1 — Full Analysis Set

+ Censored = censored observations; DFS = disease-free survival; IA1 = first interim analysis; KM = Kaplan-Meier.

Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Subgroup Analysis: DFS

The pre-specified subgroup analysis for DFS, according to the metastatic staging of patients and assessed by the investigator at baseline, is presented in the Table 32.

The following results were observed at the time of the first interim analysis:

• M0 group: A total of 102 patients out of 467 and 132 patients out of 469 had events in the pembrolizumab and placebo arms, respectively. The HR for DFS obtained from an unstratified Cox model for the pembrolizumab versus placebo was 0.74 (95% CI, 0.57 to 0.96).

• M1 NED group: A total of 7 patients out of 29 and 19 patients out of 29 had events in the pembrolizumab and placebo arms, respectively. The HR for DFS obtained between the pembrolizumab and placebo groups was 0.29 (95% CI, 0.12 to 0.69).

Similar findings were observed in the efficacy update report analysis, which had 6 additional months of follow-up, for the M0 subgroup (HR = 0.68; 95% CI, 0.53 to 0.88) and for the M1 NED subgroup (HR = 0.28; 95% CI, 0.12 to 0.66).

Post hoc analyses from efficacy update report, which reported DFS results according to recurrence risk subgroups, showed the following estimates:

• Intermediate-high risk group: A total of 87 patients out of 422 and 127 patients out of 433 had events in the pembrolizumab and placebo arms, respectively. The HR for DFS obtained for the pembrolizumab versus placebo was 0.68 (95% CI, 0.52 to 0.89).²⁷

• High-risk group: A total of 20 patients out of 40 and 23 patients out of 36 had events in the pembrolizumab and placebo arms, respectively. The HR for DFS obtained between the pembrolizumab and placebo groups was 0.60 (95% CI, 0.33 to 1.10).²⁷

• M1 NED risk group: A total of 7 patients out of 29 and 19 patients out of 29 had events in the pembrolizumab and placebo arms, respectively. The HR for DFS obtained between the pembrolizumab and placebo groups was 0.28 (95% CI, 0.12 to 0.66).²⁷

Sensitivity Analyses: DFS

Three pre-specified sensitivity analyses for DFS were reported, as per first interim analysis. In the sensitivity analysis in which disease recurrence or death documented immediately after more than 2 consecutive missed disease assessments or after initiation of new anti-cancer treatment was censored, the HR for pembrolizumab versus placebo was 0.67 (95% CI, 0.52 to 0.86; P = 0.0007) (Table 33).

Another sensitivity analysis included additional stratum for baseline disease status, based on BICR review of baseline scans, in the stratified Cox model for DFS assessment. The HR obtained for pembrolizumab versus placebo was 0.70 (95% CI, 0.54 to 0.89; P = 0.0021) (Table 34).

Participants with baseline disease, assessed by BICR, were censored in the final sensitivity analysis of DFS (presented in the Table 35). A total of 101 (20.4%) and 129 (25.9%) DFS events occurred in the pembrolizumab and placebo groups, respectively. The HR for DFS for pembrolizumab versus placebo was 0.74 (95% CI, 0.56 to 0.95; P = 0.0097).

Overall, the sensitivity analyses were consistent with the findings observed in the primary analysis of DFS and showed that the benefit of using pembrolizumab was maintained over placebo.

Disease Recurrence–Specific Survival

At the time of the first interim analysis, a total of 17 events (3.4%) of local recurrence occurred in the pembrolizumab arm and 32 events (6.4%) of local recurrence occurred in placebo arm. Distant recurrence was reported in 94 (19%) and 134 (26.9%) patients in the pembrolizumab and placebo groups, respectively. The cumulative incidence of local (DRSS1) and distant (DRSS2) recurrence, estimated with a nonparametric method adjusted for competing risks, was consistently lower in the pembrolizumab arm than in the placebo arm (Table 18, Figure 5, Figure 6).

Table 18: Disease Recurrence–Specific Survival at IA1 — Efficacy Analysis Set (ITT Population)

CI = confidence interval; DRSS = disease recurrence–specific survival; IA1 = first interim analysis; ITT = intention to treat.

Notes: Cumulative incidence estimates at specified time points are based on nonparametric estimations of cumulative incidence of the event of interest, accounting for competing risk events.

Database cut-off date: December 14, 2020.

^aLocal recurrence of RCC is counted as an event.

^bDistant kidney cancer metastasis(es) or death are counted as competing events.

^cLocal recurrence with visceral lesion or distant kidney cancer metastasis(es) with visceral lesion are counted as events.

^dDeath, local recurrence without visceral lesion, and distant metastasis without visceral lesion are counted as competing events.

Source: Clinical Study Report.¹²

Figure 5: Cumulative Incidence Plot of DRSS1 at IA1 Based on Investigator Assessment — Full Analysis Set

DRSS = disease recurrence–specific survival; IA1 = first interim analysis; RCC = renal cell carcinoma.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Figure 6: Cumulative Incidence Plot of DRSS2 at IA1 Based on Investigator Assessment — Full Analysis Set

DRSS = disease recurrence–specific survival.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Event-Free Survival

According to the findings at the first interim analysis, 116 events (23.4%) occurred in pembrolizumab arm, of which 95 were disease recurrence, 14 were disease progression, and 7 were deaths. In the placebo arm, 155 events (31.1%) occurred, of which 128 were disease recurrence, 25 were disease progression, and 2 were deaths. Median EFS was not reached in either group. HR for the comparison of EFS by BICR, with baseline disease status based only on baseline scans, in the pembrolizumab versus placebo groups was 0.72 (95% CI, 0.56 to 0.91; P = 0.0035) (Table 19, Figure 7).

Table 19: Event-Free Survival at IA1 Based on BICR (Baseline Disease Status Based on BICR Review of Baseline Scan Only) — Efficacy Analysis Set (ITT Population)

BICR = blinded independent central review; CI = confidence interval; EFS = event-free survival; HR = hazard ratio; IA1 = first interim analysis; ITT = intention to treat; NR = not reached; Q1 = first quartile; Q3 = third quartile.

Notes: For participants who were assessed as baseline NED by BICR of baseline scan only, but who had a postbaseline scan that triggered retrospective assessment of the baseline disease, the date of that scan is used as the event date.

Database cut-off date: December 14, 2020.

^aFrom product-limit (KM) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by baseline disease status by BICR (NED by BICR vs. non-NED by BICR), then within the NED by BICR group, further stratified by randomization strata: M0 vs. M1 NED by investigator and ECOG PS (0 vs. 1), US participant (yes vs. no) within M0 group by investigator.

^cOne-sided P value based on log-rank test stratified by baseline disease status by BICR (NED by BICR vs. non-NED by BICR), then within the NED by BICR group, further stratified by randomization strata: M0 vs. M1 NED by investigator and ECOG PS (0 vs. 1), US participant (yes vs. no) within M0 group by investigator.

Source: Clinical Study Report.¹²

Figure 7: Kaplan-Meier Plot of EFS at IA1 — Full Analysis Set

+ Censored = censored observations; EFS = event-free survival; IA1 = first interim analysis; KM = Kaplan-Meier

Notes: For participants who were assessed as baseline NED by BICR of baseline scan only, but who had a postbaseline scan that triggered retrospective assessment of the baseline disease, the date of that scan is used as the event date.

Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Results of the assessment of EFS based on baseline and postbaseline scans are presented in the Appendix 3.

Health-Related Quality of Life

HRQoL outcomes (FKSI-DRS, EORTC QLQ-C30 global health status/QoL and physical functioning) of the first interim analysis, presented as the overall least squares mean difference estimated for pembrolizumab versus placebo, are reported in Table 20. Scores on HRQoL measures decreased from baseline for patients in both treatment arms at week 52 (Table 20). The between-group least squares mean differences were –0.67 points (95% CI, –1.23 to –0.12; P = 0.0170) for FKSI-DRS, –2.57 points (95% CI, –5.22 to 0.08; P = 0.0571) for global health status/QoL of the EORTC QLQ-C30, and –0.91 points (95% CI, –2.79 to 0.97; P = 0.3410) for physical functioning of the EORTC QLQ-C30.

Overall least squares mean differences for the exploratory HRQoL outcomes were as follows: –0.10 points (95% CI, –1.57 to 1.37) for the nausea and vomiting symptoms of the EORTC QLQ-C30, 1.11 points (95% CI, –1.26 to 3.49) for the diarrhea symptoms of the EORTC QLQ-C30, and –1.58 points (95% CI, –3.59 to 0.42) for the EQ-5D-5L VAS (Table 39).

Appendix 3 contains the model-based least squares mean changes in EORTC QLQ-C30 global health status/QoL, functioning, and symptom scales from baseline to 52 weeks postbaseline.

The percentage of participants with improvement and improvement plus stability in FKSI-DRS and EORTC QLQ-C30 (global health status and physical functioning) scores were, overall, lower in the pembrolizumab group than in the placebo group, whereas the percentage of participants experiencing deterioration was higher in the pembrolizumab group than in the placebo group (Table 41). Differences in the percentage of participants with improvement plus stability was –9.2 (95% CI, –15.2 to –3.1) for the FKSI-DRS score, –13.6 (95% CI, –19.7 to –7.5) for the EORTC QLQ-C30 global health status, and –4.5 (95% CI, –10.3 to 1.4) for EORTC QLQ-C30 physical functioning.

Table 20: Analysis of Change From Baseline to Week 52 in PROs, Defined as Secondary End Points in the KEYNOTE-564 Trial (FKSI-DRS and EORTC QLQ-C30) — PRO FAS Population (IA1)

CFB = change from baseline; CI = confidence interval; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FKSI-DRS = Functional Assessment of Cancer Therapy Kidney Symptom Index–Disease-Related Symptoms; IA1 = first interim analysis; LS = least squares; LSM = least squares mean; QoL = quality of life; SD = standard deviation; vs. = versus.

Notes: For baseline and week 52, N is the number of patients in each treatment group with non-missing assessments at the specific time point; for change from baseline, N is the number of patients in the analysis population in each treatment group.

Database cut-off date: December 14, 2020.

^aBased on a cLDA model with PRO scores as the response variable with treatment by study visit interaction, stratification factors of metastasis status (M0 vs. M1 NED), and within the M0 group further stratified by ECOG PS (0 vs. 1) and US participant (yes vs. no) as covariates.

Source: Clinical Study Report.¹²

Harms

Only those harms identified in the review protocol are reported here. Refer to Table 21, Table 22, Table 23, Table 24, and Table 25. for detailed harms data.

Adverse Events

A total of 96.3% of patients in the pembrolizumab group and 91.1% of patients in the placebo group experienced at least 1 AE (Table 21). The most common AEs were fatigue (29.7% in pembrolizumab versus 24.2% in placebo arm), diarrhea (25.4% in pembrolizumab versus 22.4% in placebo arm), pruritus (22.7% in pembrolizumab versus 13.1% in placebo arm), arthralgia (22.1% in pembrolizumab versus 18.8% in placebo arm), hypothyroidism (21.1% in pembrolizumab versus 3.6% in placebo arm), and rash (20.1% in pembrolizumab versus 10.7% in placebo arm). Table 22 summarizes the AEs occurring in at least 10% of patients receiving pembrolizumab or placebo in the KEYNOTE-564 trial. Figure 16 provides risk differences and 95% CIs for between-group comparisons of events occurring in at least 10% of patients in either arm.

Table 21: Summary of AEs in the KEYNOTE-564 Trial (APaT Population; IA1)

AE = adverse event; APaT = all participants as treated; IA1 = first interim analysis; SAE = serious adverse event.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Table 22: Summary of AEs in the KEYNOTE-564 Trial (Incidence ≥ 10% in 1 or More Treatment Groups; APaT Population) — IA1

AE = adverse event; APaT = all participants as treated; IA1 = first interim analysis.

Notes: Every participant is counted a single time for each applicable row and column. A specific AE appears only if its incidence in at least 1 column meets the incidence criterion in the title, after rounding.

Non-SAEs up to 30 days and SAEs up to 90 days following the last dose in the initial treatment phase are included. The following MedDRA preferred terms not related to the drug have been excluded: neoplasm progression, malignant neoplasm progression, and disease progression.

Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Serious Adverse Events

A total of 20.5% and 11.3% of individuals experienced at least 1 SAE in the pembrolizumab and placebo arms, respectively (Table 23). The most reported SAEs (occurring in ≥ 1% of patients in 1 or more treatment arm) were acute kidney injury, adrenal insufficiency, pneumonia, colitis, and diabetic ketoacidosis. SAEs of grade 3 or higher were more frequently reported in the pembrolizumab arm compared with the placebo arm (32.4% versus 17.7%) (Table 23).

Table 23: Summary of Grade 3 to 5 AEs and SAEs up to 90 Days From Last Dose of Treatment (Incidence ≥ 1% in 1 or More Treatment Groups; APaT Population) — IA1

AE = adverse event; APaT = all participants as treated; IA1 = first interim analysis; SAE = serious adverse event.

Notes: Every participant is counted a single time for each applicable row and column. A specific AE appears only if its incidence in 1 or more of the columns meets the incidence criterion in the report title, after rounding. The following MedDRA preferred terms not related to the drug have been excluded: neoplasm progression, malignant neoplasm progression, and disease progression.

Database cut-off date: December 14, 2020.

^aNon-SAEs up to 30 days and SAEs up to 90 days following the last dose in the initial treatment phase are included.

^bSAEs up to 90 days after last dose of the initial treatment phase are included.

Source: Clinical Study Report.¹²

Withdrawals Due to AEs

Discontinuations Due to AEs

All-cause AEs leading to study discontinuation were reported in 20.7% of patients receiving pembrolizumab and 2.0% receiving placebo. The most common AEs resulting in treatment discontinuation included increased levels of alanine aminotransferase (1.6%), adrenal insufficiency (1.0%) and colitis (1.0%) in pembrolizumab arm.

Treatment Interruption Due to AEs

More patients (25.8%) receiving pembrolizumab experienced AEs leading to treatment interruption compared with those receiving placebo (14.9%). The most common AEs leading to the interruption of pembrolizumab were increased aspartate aminotransferase (2.3%), increased alanine aminotransferase (1.4%), hypothyroidism (1.6%), diarrhea (1.4%), fatigue (1.4%), arthralgia (1.6%), vomiting (1.0%), decreased appetite (1.0%), and rash (1.0%) in the pembrolizumab group. In the placebo group, the most frequent AEs resulting in interruption included nasopharyngitis (1.2%), diarrhea (1.0%), increased alanine aminotransferase (1.0%), and increased blood creatinine (1.0%) in the placebo group.

Mortality

At the time of the first interim analysis, 2 patients in pembrolizumab arm (0.4%) and 1 patient in placebo arm (0.2%) died due to an AE, none of which were considered treatment-related by the investigator (Table 24). In the pembrolizumab arm, 1 death was due to multiple organ dysfunction syndrome and the other was due to pneumonia; in the placebo arm, the 1 death was due to intracranial hemorrhage.

Table 24: Summary of AEs Resulting in Death up to 90 Days From Last Dose of Treatment (Incidence > 0% in 1 or More Treatment Groups; APaT Population) — IA1

AE = adverse event; APaT = all participants as treated; IA1 = first interim analysis.

Notes: Every participant is counted a single time for each applicable row and column. A specific AE appears only if its incidence in 1 or more columns meets the incidence criterion in the report title, after rounding. Non-SAEs up to 30 days and SAEs up to 90 days after the last dose in the initial treatment phase are included. The following MedDRA preferred terms not related to the drug have been excluded: neoplasm progression, malignant neoplasm progression, and disease progression.

Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Notable Harms

Hyperthyroidism occurred in 21.1% of patients treated with pembrolizumab and 6.9% of patients treated with placebo, and hypothyroidism occurred in 11.9% of patients treated with pembrolizumab and 0.2% of patients treated with placebo (Table 25). Other notable harms reported in at least 1% of patients in either treatment arm were pneumonitis (2.3% versus 1%), adrenal insufficiency (2% versus 0.2%), type 1 diabetes mellitus (1.8% versus 0%), colitis (1.6% versus 0.2%), severe skin reactions (1.6% versus 0.4%), infusion reactions (1.4% versus 1%), thyroiditis (1.2% versus 0.2%), and hepatitis (1% versus 0%).

Table 25: Summary of Notable Harms in the KEYNOTE-564 Trial (Incidence > 0% in 1 or More Treatment Groups; APaT Population) — IA1

AE = adverse event; APaT = all participants as treated; IA1 = first interim analysis.

Notes: Every participant is counted a single time for each applicable row and column. Non-SAEs up to 30 days and SAEs up to 90 days after the last dose in the initial treatment phase are included. The following MedDRA preferred terms not related to the drug have been excluded: neoplasm progression, malignant neoplasm progression, and disease progression.

Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Critical Appraisal

Internal Validity

The KEYNOTE-564 trial is an ongoing, multicentre, randomized, placebo-controlled, double-blind study. Treatment allocation was performed centrally through an Interactive Voice Response Systems and Interactive Web Response Systems. The study randomization was stratified first by patients’ metastasis status (M0 versus M1 NED). In the M0 group, ECOG PS (0 or 1) and US participant (yes or no) were additional stratification factors.

Overall, the baseline and demographic characteristics were balanced in the 2 study arms, suggesting that randomization was successfully implemented. A relatively small proportion of patients did not receive their assigned treatment (1.6% in the pembrolizumab arm and 0.4% in placebo arm), which is unlikely to have a significant impact on the comparative efficacy assessment in the ITT analysis.

Based on the study protocol, patients in the KEYNOTE-564 trial were prohibited from receiving antineoplastic systemic chemotherapy or biologic therapy, and immunotherapy during the treatment phase. However, 10 individuals (7 in the pembrolizumab arm and 3 in the placebo arm) received antineoplastic agents as concomitant medications during the trial. A small number of patients received concomitant aflibercept and fluorouracil. The clinical experts consulted by CADTH noted that these drugs are not commonly used in practice for the treatment of RCC but suggested that these patients might have experienced another type of malignancy for which they would have required chemotherapy. Even though the reasoning for concomitant anti-cancer therapy was not clearly described in the Clinical Study Report for KEYNOTE-564, it was unlikely that these concomitant therapies modified the treatment effects of pembrolizumab or had an impact on outcomes, given that they are not expected to be efficacious in the adjuvant treatment of RCC. The clinical experts consulted by CADTH agreed that the differential use of these concomitant therapies was not likely to have a meaningful effect on outcomes.

DFS and OS were the primary and the main secondary outcomes in the KEYNOTE-564 trial, respectively. Both outcomes were considered clinically relevant for patients with RCC in the adjuvant setting, according to the clinical experts consulted and the patient input provided. DFS, assessed by the investigator, was defined as the time from randomization to the first documented local recurrence, occurrence of distant kidney cancer metastasis(es), or death due to any cause, whichever occurs first. Assessment of DFS required radiologic evidence of disease recurrence, and pathologic assessment was strongly encouraged. To evaluate the robustness of the DFS findings, the sponsor introduced BICR assessments in the pivotal trial. Results showed that there was a discrepancy in the level of concordance between the investigators and BICR when it came to disease recurrence assessments. Among participants with disease recurrence, BICR agreed with investigator assessment in 81.6% of cases in the pembrolizumab group and in 86.6% in the placebo group. Conversely, among the participants without disease recurrence, BICR agreed with investigator assessment in 92.6% of cases in the pembrolizumab group and in 92.2% in the placebo group (Table 37). Furthermore, the pivotal trial provided evidence of early and late discrepancy rates, which allowed assessment of the frequency in which local investigators declared recurrence earlier or later than the BICR in each treatment group. Early and late discrepancy rates did not cross the threshold values identified in the literature that would indicate the presence of evaluation bias in favour of pembrolizumab (Table 37).²⁸ As described by the sponsor, discrepancies in disease recurrence assessments observed might have been related to the fact that disease recurrence assessed by investigators required radiographic evidence, with strong encouragement for pathology assessments, whereas BICR assessed recurrence based on radiographic evidence only. Among the 39 individuals (19 in the pembrolizumab group and 20 in the placebo group) for whom disease recurrence was declared by investigator but not by BICR, 17.9% of disease recurrences (pembrolizumab versus placebo: 15.8% versus 20.0%) were declared by the investigators based on both radiographic and pathological evidence. Moreover, among the 55 patients (28 in the pembrolizumab group and 27 in the placebo group) in whom there was no disease recurrence according to the investigators but not BICR, lack of recurrence was declared by investigators on the basis of both radiographic and pathologic evidence in 9.1% of individuals (pembrolizumab versus placebo: 14.3% versus 3.7%) (Table 38). To minimize the risk of differential measurement error, a sensitivity analysis of DFS assessment by BICR, in which participants enrolled with baseline disease assessed by BICR were censored at their randomization date, was pre-specified and conducted in the pivotal trial (Appendix 3). The sensitivity analysis suggested consistent findings with the primary DFS analysis. The clinical experts consulted by CADTH acknowledged that discrepancies in disease assessments are common in large clinical trials and noted that discrepancies in the KEYNOTE-564 trial might have been related to the fact that BICR only had radiographic information, with no other clinical information, for the trial participants. There was no clear evidence that DFS assessment by the investigator introduced bias in the results of the primary outcome evaluation.

OS was defined as the time from randomization to death due to any cause. It is important to note that surgery in patients with RCC is performed with curative intent, and that the disease-specific survival period postnephrectomy is rather long in patients with intermediate- or high-risk disease (i.e., 5-year rates are about 80% for individuals at intermediate risk and 40% to 55% for individuals at high risk).³ Thus, a longer follow-up is likely required to observe the effect of treatment on survival outcomes in the adjuvant setting, which was evidenced by the fact that median DFS and OS were not reached in the pivotal study, and was further validated by the clinical experts consulted for this CADTH review. In the KEYNOTE-564 study, treatment was limited to 17 doses (or approximately 1 year of treatment), and patients could receive subsequent systemic anti-cancer regimens (e.g., radiation, surgery, medication) after discontinuing their study treatment. Specifically, more patients in the placebo arm (22.5%) received post-treatment anti-cancer therapies compared to the pembrolizumab arm (15.3%). An increased use of anti-cancer therapies among patients in the placebo group would introduce bias in the comparative assessment of OS and would make it harder to demonstrate between-group differences in OS during the trial period (i.e., the direction of bias would likely go against pembrolizumab). Furthermore, subsequent treatments also make it difficult to determine the longer-term benefits of adjuvant pembrolizumab on OS. CADTH reviewers therefore deemed that it is difficult to determine the benefits of pembrolizumab on improving survival in the adjuvant setting, based on the current trial evidence.

Sample size and power calculations were based on DFS and OS. When calculating sample size and power of the KEYNOTE-564 trial, the sponsor assumed that DFS would follow a Poisson mixture cure rate model with a cure rate of 0.3 in the long term, based on previously published data.²⁵ However, the study protocol did not specify which recurrence risk groups this estimate applied to (intermediate-high, high, or both) or the time frame the estimate was applied to. The clinical experts consulted by CADTH agreed with the reviewers that these strata would affect the cure rate used. Nevertheless, the clinical experts acknowledged that the assumptions are consistent with a cure rate averaged across the recurrence risk strata and noted that it seemed reasonable for the a priori estimation of sample size for a trial.

The Cox proportional hazards model, used for the assessment of primary and secondary study outcomes, assumes that the hazards are proportional in the 2 treatment arms. Because a violation of this assumption may lead to biased estimates in the regression model, the sponsor’s statistical analysis plan pre-specified that proportional hazards assumptions related to DFS and OS would be assessed visually and analytically, if warranted. If that visual inspection yielded evidence of nonproportional hazards, the sponsor pre-specified possible supportive analyses using restricted mean survival time and parametric methods.^29,30 Survival curves were estimated using the KM model. CADTH’s visual assessment of the DFS curve did not provide evidence suggestive of any violation in proportional hazards.

The amounts of censored data across the efficacy outcomes were fairly balanced between the pembrolizumab and placebo study groups. However, censoring in the study might have been affected by imbalances in treatment discontinuation due to AEs (38.9% in the pembrolizumab arm; 26.2% in the placebo arm), as well as by potential unblinding of the study due to irAEs (35.5% in the pembrolizumab arm; 6.9% in the placebo arm). The sponsor did not pre-specify a sensitivity analysis that would consider censoring related to AEs. Still, a conservative approach was adopted to handle treatment discontinuation (i.e., patients who completed the study treatment or discontinued treatment due to a reason other than disease recurrence were considered to be on study and followed scheduled efficacy assessments). In addition, the outcomes of DFS and EFS analyses by BICR showed findings consistent with the primary analysis. CADTH reviewers considered that the potential for informative censoring due to treatment discontinuation related to AEs and the potential for unblinding in the KEYNOTE-564 trial were minimal.

HRQoL was assessed as a secondary outcome in the KEYNOTE-564 trial using the FKSI-DRS and EORTC QLQ-C30 (global health status/QoL and physical functioning) questionnaires. The EQ-5D-5L VAS was assessed as an exploratory outcome in the trial. None of the HRQoL outcomes were controlled for multiplicity. The clinical experts consulted by CADTH noted that the questionnaires used in the KEYNOTE-564 trial were appropriate for assessing PROs in patients with RCC in the adjuvant setting. The validity of the FKSI-DRS questionnaire has been documented in patients with advanced and metastatic RCC, with evidence of reliability and responsiveness, but no evidence was found for patients in the adjuvant setting. Both EORTC QLQ-C30 and EQ-5D-5L have been widely used in oncologic trials in different cancer populations, but the validity of these questionnaires has not been documented in patients with RCC in the adjuvant setting. Similarly, no MID was established for EORTC QLQ-C30 and EQ-5D-5L measures in the population of interest for this review. Statistical analysis of PROs included a cLDA model, which is best suited for data that are missing at random or missing completely at random. In the current analysis, estimates from this model might have been biased, as there is a high likelihood that data in the KEYNOTE-564 trial are not missing at random. It was not reported whether the sensitivity analysis was performed to investigate the impact of the missing data or whether the missing-at-random assumption was met. Interpretation of the HRQoL results is more difficult with missing outcome data at later time points, as evidenced by the decline in rates of completed questionnaires to approximately 60% at week 52. According to the study protocol, the handling of missing data was model-based. However, the exact structure of the model was not specified, and the appropriateness of the model could not be determined.

Regarding the multiplicity adjustments, the type I error rate was controlled with a fixed testing sequence (DFS at a 1-sided alpha level of 2.5% first, and if DFS was successful, the alpha was then passed to OS). Sensitivity analyses and adjustments of covariates were conducted for DFS, and the findings were consistent with the primary analysis in the ITT set. Multiplicity adjustments were not conducted for other secondary or exploratory end points (DRSS, EFS, and HRQoL), including the analysis of subgroups.

A set of subgroup analyses was pre-specified in the protocol. There was no evidence of formal statistical tests of interaction being performed to test whether treatment effects differed among subgroups for the primary end point. The subgroup analysis of interest in this CADTH report (metastasis status) was based on a randomization stratification variable and was not adjusted for multiplicity. The small sample size in the M1 NED group suggests that this analysis might be underpowered to detect a meaningful difference between the study arms. Moreover, an additional subgroup analysis of DFS according to the recurrence risk groups was provided at the efficacy update report analysis but was not pre-specified in the protocol. Therefore, the study was not adequately designed to evaluate the effects of adjuvant pembrolizumab across all relevant subgroups.

The results reported in the Clinical Study Report for KEYNOTE-564 were based on the first interim analysis, with a data cut-off date of December 14, 2020. The study protocol specified 3 interim and 1 final analyses, with stopping rules defined a priori. Moreover, the sponsor implemented an additional amendment to the supplemental statistical plan to accommodate potential requests from regulatory agencies (efficacy update report analysis). The efficacy update report analysis, with an additional 6 months of follow-up and a data cut-off of June 14, 2021, was added between the first and second interim analysis. The group sequential approach that was used to allocate alpha between the interim and final analyses was considered appropriate by the CADTH reviewers.

The analysis for the primary and main secondary outcomes (DFS and OS) was conducted based on the ITT principle. Overall, clinically important protocol deviations were reported for 2 participants receiving pembrolizumab and 3 participants receiving placebo at the first interim analysis. In the study analyses, no data points were excluded because of an important protocol deviation. No per protocol analysis was conducted to check for consistency with the ITT analysis, but the number of patients with important deviations was low and balanced in the 2 study groups. Hence, CADTH deemed the risk of bias owing to protocol deviations to be low, and to likely have a negligible influence on comparative efficacy findings.

External Validity

The KEYNOTE-564 study was a multinational, multicentre trial. Of the 212 sites in 21 countries that participated, there were 10 sites in Canada. The population for the requested reimbursement aligns with the approved Health Canada indication.

The clinical experts consulted by CADTH agreed that the inclusion and exclusion criteria of the trial were appropriate for the RCC population in the adjuvant setting (Table 26). The experts also noted that the baseline characteristics of patients in the KEYNOTE-564 trial could be generalized to patients in the Canadian setting. According to the experts, the proportion of patients in the different risk-of-recurrence groups was deemed to be reflective of the population they would consider for adjuvant pembrolizumab treatment in real-world practice. It is uncertain whether the findings can be generalized to patients with a histology other than clear cell and with an ECOG PS of at least 2, as no such patients were included in the study.

In terms of the frequency of disease assessment, the clinical experts anticipated that disease assessments would be conducted less frequently in real-world practice than they were in the trial. Concomitant medications administered during the trial and subsequent anti-cancer therapies administered during follow-up were considered appropriate by the clinical experts consulted and reflective of medications administered in the Canadian setting.

In the KEYNOTE-564 trial, the only administered dosage of pembrolizumab was 200 mg every 3 weeks for up to 17 doses. The clinical experts indicated that physicians would prefer a dosing of 400 mg every 6 weeks for up to 9 doses in real-world clinical practice. Furthermore, the experts reported that weight-based dosing has been introduced in some Canadian provinces to optimize the costs and decrease drug wastage.

Placebo was considered an appropriate comparator because there are no Health Canada–approved adjuvant treatment options available in the Canadian setting.

The clinical experts consulted during this CADTH review highlighted that OS, DFS, HRQoL, and safety are appropriate outcomes to consider for patients with RCC in the adjuvant setting.

Table 26: Assessment of the Generalizability of Evidence for Pembrolizumab

AE = adverse event; APat = all participants as treated; DFS = disease-free survival; DRSS = disease recurrence–specific survival; ECOG PS = Eastern Cooperative Oncology Group Performance Status; EFS = event-free survival; FAS = full analysis set; HRQoL = health-related quality of life; ITT = intention to treat; IV = IV; M0 = no distant metastases; M1 NED = solid, isolated, soft tissue metastases that are either synchronous (i.e., completely resected at the time of nephrectomy) or metachronous (i.e., completely resected ≤ 1 year from nephrectomy); OS = overall survival; RCC = renal cell carcinoma.

Indirect Evidence

Through screening the articles yielded by the literature search, CADTH reviewers identified a systematic review and a network meta-analysis of safety and efficacy outcomes that compared adjuvant pembrolizumab with adjuvant TKIs in patients with nonmetastatic RCC postnephrectomy.³¹ This publication was ultimately excluded from the current review because the comparators were deemed irrelevant for the Canadian context.

Other Relevant Evidence

No other relevant study was identified in this review.

Discussion

Summary of Available Evidence

This CADTH systematic review included evidence from 1 pivotal study (KEYNOTE-564) and considered additional input from patient groups, clinical experts, clinician groups, and drug plans.

The KEYNOTE-564 trial is an ongoing multicentre, randomized, double-blind, phase III study comparing the efficacy and safety of pembrolizumab with placebo as an adjuvant treatment for adult patients with RCC postnephrectomy or postnephrectomy and resection of metastatic lesions. The study enrolled patients aged 18 years and older with a histologically confirmed diagnosis of RCC with a clear cell component, with or without sarcomatoid features. The study included patients at intermediate-high risk (pT2, grade 4 or sarcomatoid, N0, M0; or pT3, any grade, N0, M0), or high risk (pT4, any grade, N0, M0; or pT any stage, any grade, N+, M0) of recurrence, and M1 NED patients (patients with a primary kidney tumour and solid, isolated, soft tissue metastases completely resected either at the time of [synchronous] or 1 year or less after nephrectomy [metachronous]).

Patients were randomized in a 1:1 ratio to 1 of 2 treatment arms (pembrolizumab or placebo) based on following stratification factors: metastasis status variable (M0 versus M1 NED) and, in the M0 group, ECOG PS (0 or 1) and US participant (yes or no). The primary outcome of the KEYNOTE-564 trial was DFS, assessed by the investigator, and the main secondary outcome was OS. Other secondary and exploratory outcomes investigated included EFS, DRSS, HRQoL, and safety.¹²

Patients in the KEYNOTE-564 trial received either pembrolizumab (200 mg IV infusion every 3 weeks) or placebo (saline solution IV every 3 weeks) for up to a maximum of 17 infusions (or approximately 1 year), or until confirmation of recurrence, treatment discontinuation, or study termination. At the first interim analysis (December 14, 2020), 994 patients were randomized into the trial (496 in the pembrolizumab arm and 488 in the placebo arm; 10 patients did not receive their assigned treatment). The median age of the patients enrolled in the study was 60 years, and majority of participants were male and White. Most patients had tumours with an absence of sarcomatoid features and were at intermediate-high risk of recurrence. Baseline characteristics were balanced in the 2 study arms.

No indirect treatment comparisons or other relevant studies were identified for this review.

Interpretation of Results

Efficacy

Both the clinical experts and clinician groups consulted during the CADTH review highlighted that lengthening survival (OS) and reducing the risk of recurrence (DFS), without negatively influencing patients’ HRQoL are considered important treatment goals for patients with RCC in the adjuvant setting. Patient group input highlighted the need for adjuvant treatment options that would reduce the risk of disease recurrence after nephrectomy. Specifically, the input collected from patient groups by CADTH revealed that patients experience distress from the expectation of recurrence, and that almost half of the patients would be willing to accept adjuvant immunotherapy postsurgery if the therapy reduced the risk of disease recurrence by 40% to 50%.

The CADTH review protocol identified OS, DFS, EFS, DRSS, HRQoL, and safety as important outcomes for patients. All these outcomes were pre-specified in the KEYNOTE-564 trial before the first interim data cut-off (December 14, 2020), and the type I error rate was adequately controlled for OS and DFS. The CADTH review identified 3 subgroups of interest (histology, recurrence risk, and metastatic staging). The KEYNOTE-564 trial pre-specified metastatic staging as a subgroup analysis for DFS, which was also a stratification variable used in the randomization. Other subgroup analysis of DFS by recurrence risk groups was provided at the efficacy update report analysis, which had an additional 6 months of follow-up but was not pre-specified in the protocol. None of the subgroup analyses accounted for multiplicity. Sensitivity analyses were conducted for DFS, and the findings obtained were consistent with the primary DFS findings. Analyses of other outcomes, such as the EFS, DRSS, HRQoL, and safety, were not adjusted for multiplicity in the analysis.

The primary outcome was statistically significant at the first interim analysis data cut-off. Median DFS was not reached in either study group. The HR for DFS for pembrolizumab versus placebo was 0.68 (95% CI, 0.53 to 0.87; P = 0.001). The DFS rates at 24 months were 77.3% (95% Cl, 72.8% to 81.1%) in the pembrolizumab arm and 68.1% (95% Cl, 63.5% to 72.2%) in the placebo arm, suggesting a 9.2% relative difference. Findings from the efficacy update report analysis were consistent with results presented at first interim analysis. The clinical experts acknowledged that the DFS results were clinically meaningful.

The pre-specified DFS subgroup analysis according to the metastatic staging revealed a more pronounced effect in the M1 group (HR = 0.29; 95% CI, 0.12 to 0.69) than in the M0 group (HR = 0.74; 95% CI, 0.57 to 0.96) at the first interim analysis. These findings were consistent with data from the additional 6-months of follow-up presented in the efficacy update report analysis. The efficacy update report post hoc analysis of DFS by recurrence risk group showed a HR of 0.68 (95% CI, 0.52 to 0.89), 0.60 (95% CI, 0.33 to 1.10), and 0.28 (95% CI, 0.12 to 0.66) for the intermediate-high, high, and M1 NED risk groups, respectively. The findings from the analysis of subgroups were considered exploratory, as sample sizes were relatively small in certain subgroups and no multiplicity adjustments were used in these analyses.

Median OS had not been reached at the time of first interim analysis (December 14, 2020) or efficacy update report (June 14, 2021) analyses. The HR estimates for pembrolizumab versus placebo were 0.54 (95% CI, 0.30 to 0.96; P = 0.0164) at the first interim analysis and 0.52 (95% CI, 0.31 to 0.86; P = 0.005) at the efficacy update report analysis. Surgery in patients with RCC is performed with curative intent, and 5-year disease-specific survival is relatively long in patients at intermediate (about 80%) and high risk (from 40% to 55%) of recurrence postnephrectomy.³ The median duration of follow-up at the first interim analysis was approximately 24 months in the 2 treatment groups. In consultation with the clinical experts for CADTH, OS data from the pivotal KEYNOTE-564 trial were considered immature, owing to the shorter follow-up time relative to the natural history of patients in the adjuvant setting with RCC. Moreover, the fixed treatment duration recommended for pembrolizumab (median on-treatment exposure in the study was slightly longer than 11 months), with the opportunity for patients in either group to receive subsequent anti-cancer treatments, makes it difficult to determine the true effect of pembrolizumab on OS. Of note, there was a higher proportion of individuals receiving post-treatment anti-cancer therapies in the placebo than in the pembrolizumab group, which might have introduced bias against pembrolizumab in the assessment of OS. Because of these issues, the benefit of adjuvant pembrolizumab for OS in patients with RCC postnephrectomy or postnephrectomy and resection of metastatic lesions remains uncertain. Longer follow-up is likely required to observe the effects of adjuvant treatment on survival. Interim analyses 2 and 3 as well as the final OS analysis are planned when a pre-specified number of events have accrued in the pivotal trial (N = 132; N = 172, and N = 200, respectively).

Through the discussions with the clinical experts consulted for this review, the CADTH team collected feedback on the importance of OS as an outcome for patients with RCC in the adjuvant setting. It is important to note that evidence of the relationship between OS and its surrogate end point (DFS) is uncertain and is moderate, according to the limited data published in the literature (correlations ranging from 0.44 to 0.7).^32,33

The results of EFS and the EFS sensitivity analysis described in the KEYNOTE-564 trial were, overall, aligned with results from the DFS primary ITT analysis. Cumulative incidence rates of disease recurrence were consistently lower in the pembrolizumab arm compared with the placebo arm for both local (DRSS1) and distant (DRSS2) recurrence events. However, these outcomes were not adjusted for multiplicity and were considered to be of lower clinical importance as described by the clinical experts consulted for this CADTH review.

The clinician and patient groups consulted during the CADTH review identified patient-reported and HRQoL outcomes as important treatment goals. In the KEYNOTE-564 trial, HRQoL was assessed using 3 questionnaires. The FKSI-DRS has been validated in patients with advanced and metastatic RCC, with MIDs ranging from 0.62 to 3 points, depending on the different anchors. No studies assessing validity and MIDs were found for the adjuvant setting. No studies assessing the validity, reliability, and responsiveness of EORTC QLQ-C30 (secondary outcome) and EQ-5D-5L (exploratory outcome) in patients with RCC in the adjuvant setting have been identified in published literature. The clinical experts highlighted the absence of HRQoL literature in the adjuvant setting and noted that the questionnaires in the KEYNOTE-564 trial may be considered appropriate for this RCC patient population. The clinical experts emphasized the importance of a patient’s quality of life not deteriorating during adjuvant treatment and reported that differences in HRQoL measures between the pembrolizumab and placebo arms did not seem substantial. Despite the higher proportion of irAEs in the pembrolizumab group, the clinicians noted that changes in functioning in the pembrolizumab group were comparable to those observed in the placebo group. The rates of completed questionnaires declined to about 60% at week 52 in both treatment arms. It was unclear whether the cLDA model used for the analysis of HRQoL outcomes adequately accounted for missing data, given the limited information provided. Moreover, HRQoL analyses were exploratory and did not adjust for multiplicity, so the results presented in the pivotal trial were considered descriptive by CADTH. The impact of adjuvant pembrolizumab on HRQoL, compared with placebo, is uncertain, given the lack of formally established MIDs and the limitations identified in the analyses.

According to the clinical experts consulted by CADTH, patients recruited in the trial were considered to be representative of patients in Canadian clinical practice. The distribution of patients in the different risk-of-recurrence groups was considered to be reflective of the population that might be considered for adjuvant pembrolizumab treatment in clinical practice, according to the experts. There were no major concerns about the generalizability of the findings to Canadian practice. The clinical experts acknowledged that it is uncertain whether the findings can be generalized to patients with histologies other than clear cell or to patients with an ECOG PS of 2 or higher, as no such patients were included in the pivotal trial.

Harms

Overall, the proportion of patients with at least 1 AE was higher in the pembrolizumab group (96.3%) than in the placebo group (91.1%). The frequency of SAEs was also higher in the pembrolizumab arm (20.5%) than in the placebo arm (11.3%) in the KEYNOTE-564 study. There were more AEs leading to drug discontinuations (pembrolizumab versus placebo: 20.7% versus 2.0%) and treatment interruptions (25.8% versus 14.9%) in the pembrolizumab arm compared to placebo. Two deaths were reported in the pembrolizumab arm (0.4%), and 1 death was reported in the placebo arm (0.2%), none of which were considered to be related to treatment by the investigator. Notable harms were more frequently reported in the pembrolizumab group compared with the placebo group. The higher incidence of notable harms in the pembrolizumab group than in the placebo group was driven by hyperthyroidism (pembrolizumab versus placebo: 21.1% versus 6.9%), hypothyroidism (11.9% versus 0.2%), pneumonitis (2.3% versus 1.0%), adrenal insufficiency (2.0% versus 0.2%), type 1 diabetes mellitus (1.8% versus 0.0%), colitis (1.6% versus 0.2%), severe skin reactions (1.6% versus 0.4%), infusion reactions (1.4% versus 1.0%), thyroiditis (1.2% versus 0.2%), and hepatitis (1.0% versus 0.0%).

Input from patient groups received for this review reported that approximately half of patients would be willing to accept the risk of side effects associated with steroid use to manage the side effects of adjuvant immunotherapy, if that level of risk was in the range of 20% to 25%.

Overall, the clinical experts consulted by CADTH for this review reported that the safety profile of pembrolizumab observed in this study appeared to be consistent with the known safety profile of immune-oncologic therapy, and no additional safety signals were identified. Moreover, the clinical experts indicated that irAEs should not be underestimated, as some of them are irreversible conditions that can be life-altering for the patients (e.g., type 1 diabetes, pneumonitis). Finally, the experts highlighted the importance of frequent and close monitoring for treatment-related AEs.

Conclusions

One sponsor-submitted, multicentre, randomized, double-blind, phase III trial (KEYNOTE-564) was included in this CADTH systematic review.

Overall, pembrolizumab improved DFS outcome, compared with placebo, as an adjuvant treatment for patients with RCC who are at intermediate-high or high risk of recurrence after nephrectomy or following nephrectomy and resection of metastatic lesions. However, the effects of adjuvant pembrolizumab relative to placebo on OS could not be determined because of the immature survival data, uncertain influence of subsequent treatments, and uncertainty about the correlation between DFS and OS in the adjuvant treatment of RCC. Likewise, limitations in the HRQoL analyses in the single randomized controlled trial precluded the drawing of conclusions about the effects of pembrolizumab on this outcome. The safety profile of pembrolizumab was similar to that observed in other trials of this drug, including effects on the thyroid and adrenal glands. The clinical experts considered the baseline characteristics and the findings from the KEYNOTE-564 trial to be generalizable to patients with RCC in the adjuvant setting in Canada.

References

1.NCCN guidelines for patients: kidney cancer. Plymouth Meeting (PA): National Comprehensive Cancer Network; 2022: https://www.nccn.org/patients/guidelines/content/PDF/kidney-patient.pdf. Accessed 2022 May 11.

2.EAU guidelines on renal cell carcinoma. Arnhem (NL): European Association of Urology; 2022: https://uroweb.org/guidelines/renal-cell-carcinoma. Accessed 2022 May 11.

3.Renal cell carcinoma: ESMO clinical practice guidelines. Lugano (CH): European Society for Medical Oncology; 2019: https://www.esmo.org/guidelines/genitourinary-cancers/renal-cell-carcinoma Accessed 2022 May 11.

4.Canadian cancer statistics. 2021. Toronto (ON): Canadian Cancer Society; 2021: https://cdn.cancer.ca/-/media/files/research/cancer-statistics/2021-statistics/2021-pdf-en-final.pdf. Accessed 2022 May 11.

5.National Cancer Institute. Kidney (renal cell) cancer—health professional version 2022; https://www.cancer.gov/types/kidney/hp. Accessed 2022 May 11.

6.Rini BI, Campbell SC, Escudier B. Renal cell carcinoma. Lancet. 2009;373(9669):1119-1132. PubMed

7.Patard JJ, Kim HL, Lam JS, et al. Use of the University of California Los Angeles integrated staging system to predict survival in renal cell carcinoma: an international multicenter study. J Clin Oncol. 2004;22(16):3316-3322. PubMed

8.Zigeuner R, Hutterer G, Chromecki T, et al. External validation of the Mayo Clinic stage, size, grade, and necrosis (SSIGN) score for clear-cell renal cell carcinoma in a single European centre applying routine pathology. Eur Urol. 2010;57(1):102-109. PubMed

9.Renal cell carcinoma. Edmonton (AL): Cancer Care Alberta; 2021: https://www.albertahealthservices.ca/assets/info/hp/cancer/if-hp-cancer-guide-gu003-renal-cell.pdf. Accessed 2022 May 11.

10.Canil C, Kapoor A, Basappa NS, et al. Management of advanced kidney cancer: Kidney Cancer Research Network of Canada (KCRNC) consensus update 2021. Can Urol Assoc J. 2021;15(4):84-97. PubMed

11.Keytruda (pembrolizumab): powder for solution for infusion 50 mg, solution for infusion 100 mg/4 mL vial [product monograph]. Kirkland (QC): Merck Canada; 2021 Aug 13.

12.Clinical Study Report: P564V01MK3475. A phase 3, randomized, double-blind, placebo- controlled clinical trial of pembrolizumab (MK-3475) as monotherapy in the adjuvant treatment of renal cell carcinoma post nephrectomy (KEYNOTE-564) [internal sponsor's report]. Rahway (NJ): Merck Sharp & Dohme Corp; 2021 Jun 7.

13.Leibovich BC, Lohse CM, Crispen PL, et al. Histological subtype is an independent predictor of outcome for patients with renal cell carcinoma. J Urol. 2010;183(4):1309-1315. PubMed

14.BC Cancer. Kidney. 2021; http://www.bccancer.bc.ca/health-info/types-of-cancer/urinary/kidney Accessed 2022 May 11.

15.McGowan J, Sampson M, Salzwedel DM, Cogo E, Foerster V, Lefebvre C. PRESS Peer Review of Electronic Search Strategies: 2015 guideline statement. J Clin Epidemiol. 2016;75:40-46. PubMed

16.Grey matters: a practical tool for searching health-related grey literature. Ottawa (ON): CADTH; 2019: https://www.cadth.ca/grey-matters. Accessed 2022 Mar 21.

17.Choueiri TK, Tomczak P, Park SH, et al. Adjuvant pembrolizumab after nephrectomy in renal-cell carcinoma. N Engl J Med. 2021;385(8):683-694. PubMed

18.EORTC Quality of Life Group. EORTC QLQ-C30 - FAQ. 2021; https://qol.eortc.org/faq. Accessed 2021 Dec 23.

19.EORTC Quality of Life Group, Fayers P, Aaronson N, et al. The EORTC QLQ-C30 scoring manual, 3rd ed. Brussels (BE): European Organisation for Research and Treatment of Cancer; 2001: https://www.eortc.org/app/uploads/sites/2/2018/02/SCmanual.pdf. Accessed 2021 Dec 23.

20.Cella D, Yount S, Brucker PS, et al. Development and validation of a scale to measure disease-related symptoms of kidney cancer. Value Health. 2007;10(4):285-293. PubMed

21.FACIT Group. FKSI-DRS Scoring Menu. Functional Assessment of Cancer Therapy – Kidney Symptom Index – Disease Related Symptoms (FKSI-DRS). Scoring guidelines (version 4). 2021; https://www.facit.org/measures-scoring-downloads/fksi-drs-scoring-downloads. Accessed 2021 Dec 23.

22.EuroQol Research Foundation. EQ-5D-5L user guide: basic information on how to use the EQ-5D-5L instrument. 2019; https://euroqol.org/publications/user-guides/. Accessed 2022 May 11.

23.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(1):139-144. PubMed

24.Cella D, Li JZ, Cappelleri JC, et al. Quality of life in patients with metastatic renal cell carcinoma treated with sunitinib or interferon alfa: results from a phase III randomized trial. J Clin Oncol. 2008;26(22):3763-3769. PubMed

25.Leibovich BC, Blute ML, Cheville JC, et al. Prediction of progression after radical nephrectomy for patients with clear cell renal cell carcinoma: a stratification tool for prospective clinical trials. Cancer. 2003;97(7):1663-1671. PubMed

26.Clinical Study Report (efficacy update report): P564V01MK3475. 2.7.3 summary of clicnical efficacy [internal sponsor's report]. Rahway (NJ): Merck Sharp & Dohme Corp;.

27.Choueiri TK, Tomczak P, Park SE, et al. Pembrolizumab as post nephrectomy adjuvant therapy for patients with renal cell carcinoma: results from 30-month follow-up of KEYNOTE-564. J Clin Oncol. 2022;40(6 suppl):290.

28.Mannino FV, Amit O, Lahiri S. Evaluation of discordance measures in oncology studies with blinded independent central review of progression-free survival using an observational error model. J Biopharm Stat. 2013;23(5):971-985. PubMed

29.Uno H, Claggett B, Tian L, et al. Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis. J Clin Oncol. 2014;32(22):2380-2385. PubMed

30.Anderson KM. A nonproportional hazards Weibull accelerated failure time regression model. Biometrics. 1991;47(1):281-288. PubMed

31.Laukhtina E, Quhal F, Mori K, et al. Pembrolizumab outperforms tyrosine kinase inhibitors as adjuvant treatment in patients with high-risk renal cell carcinoma after nephrectomy. Eur Urol Oncol. 2022;5(1):120-124. PubMed

32.Haas NB, Song Y, Willemann Rogerio J, et al. Disease-free survival as a predictor of overall survival in localized renal cell carcinoma (RCC) following first nephrectomy. J Clin Oncol. 2021;39(15 suppl):4581.

33.Harshman LC, Xie W, Moreira RB, et al. Evaluation of disease-free survival as an intermediate metric of overall survival in patients with localized renal cell carcinoma: a trial-level meta-analysis. Cancer. 2018;124(5):925-933. PubMed

34.Capitanio U, Larcher A, Montorsi F. Re: Toni K. Choueiri, Piotr Tomczak, Se Hoon Park, et al. Adjuvant pembrolizumab after nephrectomy in renal-cell carcinoma. N Engl J Med 2021;385:683-94: Adjuvant pembrolizumab after nephrectomy: a plea to reconsider the need for lymph node dissection. Eur Urol. 2022;81(1):e28. PubMed

35.Khene ZE, Bex A, Bensalah K. Adjuvant therapy after surgical resection of nonmetastatic renal cell carcinoma: one size does not fit all. Eur Urol. 2022;81(4):432-433. PubMed

36.Magee DE, Kutikov A. Re: adjuvant pembrolizumab after nephrectomy in renal-cell carcinoma. Eur Urol. 2022;81(3):317-318. PubMed

37.Saleh K, Kordahi M, Felefly T, Kourie HR, Khalife N. Pembrolizumab: first adjuvant immunotherapy in renal cell carcinoma? Future Oncol. 2022;18(5):519-522. PubMed

38.Msaouel P, Grivas P, Zhang T. Adjuvant systemic therapies for patients with renal cell carcinoma: choosing treatment based on patient-level characteristics. Eur Urol Oncol. 2021. PubMed

39.Nierengarten MB. Pembrolizumab may be a new adjuvant treatment following renal cell carcinoma surgery. Cancer. 2021;127(20):3713. PubMed

40.Sharma A, Sahoo RK, Batra A. Adjuvant pembrolizumab after nephrectomy in renal-cell carcinoma. N Engl J Med. 2021;385(20):1919. PubMed

41.Drug Reimbursement Review sponsor submission: Keytruda (pembrolizumab), 200mg, intravenous infusion [internal sponsor's package]. Kirkland (QC): Merck Canada; 2022 Mar 3.

42.Giesinger JM, Kieffer JM, Fayers PM, et al. Replication and validation of higher order models demonstrated that a summary score for the EORTC QLQ-C30 is robust. J Clin Epidemiol. 2016;69:79-88. PubMed

43.FACIT Group. FKSI-DRS: Overview. 2021; https://www.facit.org/measures/FKSI-DRS. Accessed 2021 Dec 23.

44.Cella D, Motzer RJ, Rini BI, et al. Important group differences on the Functional Assessment of Cancer Therapy-Kidney Symptom Index Disease-Related Symptoms in patients with metastatic renal cell carcinoma. Value Health. 2018;21(12):1413-1418. PubMed

45.EuroQol--a new facility for the measurement of health-related quality of life. Health Policy. 1990;16(3):199-208. PubMed

46.Brooks R. EuroQol: the current state of play. Health Policy. 1996;37(1):53-72. PubMed

47.McClure NS, Sayah FA, Xie F, Luo N, Johnson JA. Instrument-defined estimates of the minimally important difference for EQ-5D-5L index scores. Value Health. 2017;20(4):644-650. PubMed

Appendix 1: Literature Search Strategy

Note this appendix has not been copy-edited.

Clinical Literature Search

Overview

Interface: Ovid

Databases

• MEDLINE All (1946 to present)

• Embase (1974 to present)

Note: Subject headings and search fields have been customized for each database. Duplicates between databases were removed in Ovid.

Date of search: April 1, 2022

Alerts: Bi-weekly search updates until project completion

Search filters applied: No filters were applied to limit the retrieval by study type.

Limits

• No date or language limits were used

• Conference abstracts: excluded

Table 27: Syntax Guide

Multi-Database Strategy

1. (Keytruda* or pembrolizumab* or lambrolizumab* or MK 3475 or MK3475 or Merck 3475 or HSDB 8257 or HSDB8257 or Sch 900475 or Sch900475 or DPT0O3T46P).ti,ab,kf,ot,hw,rn,nm.

2. Kidney Neoplasms/ or Carcinoma, Renal Cell/

3. ((kidney* or renal or hypernephroid or hyper-nephroid* or collecting duct* or Grawitz* or nephroid) adj3 (cancer* or carcinoma* or neoplas* or tumor* or tumour* or adenocarcinoma* or adeno-carcinoma* or pyelocarcinoma* or metast* or malignan* or sarcoma*)).ti,ab,kf.

4. (hypernephroma* or nephroma* or reninoma* or RCC or mRCC).ti,ab,kf.

5. or/2-4

6. 1 and 5

7. 6 use medall

8. *pembrolizumab/

9. (Keytruda* or pembrolizumab* or lambrolizumab* or MK 3475 or MK3475 or Merck 3475 or HSDB 8257 or HSDB8257 or Sch 900475 or Sch900475).ti,ab,kf,dq

10. 8 or 9

11. exp Kidney cancer/

12. ((kidney* or renal or hypernephroid or hyper-nephroid* or collecting duct* or Grawitz* or nephroid) adj3 (cancer* or carcinoma* or neoplas* or tumor* or tumour* or adenocarcinoma* or adeno-carcinoma* or pyelocarcinoma* or metast* or malignan* or sarcoma*)).ti,ab,kf,dq.

13. (hypernephroma* or nephroma* or reninoma* or RCC or mRCC).ti,ab,kf,dq.

14. or/11-13

15. 10 and 14

16. 15 use oemezd

17. 16 not (conference review or conference abstract).pt.

18. 7 or 17

19. remove duplicates from 18

Clinical Trials Registries

ClinicalTrials.gov

Produced by the US National Library of Medicine. Targeted search used to capture registered clinical trials.

• [Search -- Studies with results | pembrolizumab or Keytruda and renal cell carcinoma]

WHO ICTRP

International Clinical Trials Registry Platform, produced by the WHO. Targeted search used to capture registered clinical trials.

• [Search terms -- pembrolizumab or Keytruda and renal cell carcinoma]

Health Canada’s Clinical Trials Database

Produced by Health Canada. Targeted search used to capture registered clinical trials.

• [Search terms -- pembrolizumab or Keytruda and renal cell carcinoma]

EU Clinical Trials Register

European Union Clinical Trials Register, produced by the European Union. Targeted search used to capture registered clinical trials.

• [Search terms -- pembrolizumab or Keytruda and renal cell carcinoma]

Grey Literature

Search dates: March 21 to March 28, 2022

Keywords: pembrolizumab, Keytruda, renal cell carcinoma

Limits: None

Updated: Search updated before the completion of stakeholder feedback period

Relevant websites from the following sections of the CADTH grey literature checklist Grey Matters: A Practical Tool for Searching Health-Related Grey Literature were searched:

• Health Technology Assessment Agencies

• Health Economics

• Clinical Practice Guidelines

• Drug and Device Regulatory Approvals

• Advisories and Warnings

• Drug Class Reviews

• Clinical Trials Registries

• Databases (free)

• Health Statistics

• Internet Search

Appendix 2: Excluded Studies

Note this appendix has not been copy-edited.

Table 28: Excluded Studies

Appendix 3: Detailed Outcome Data

Note this appendix has not been copy-edited.

Table 29: Protocol Amendments

DFS = disease-free survival; EFS = event-free survival; IA = interim analysis; M1 NED = patients who present not only with the primary kidney tumour but also solid, isolated, soft tissue metastases that were completely resected at the time of nephrectomy (synchronous) or ≤ 1 year from nephrectomy (metachronous).

Source: Clinical Study Report.¹²

Table 30: Analysis of Overall Survival in the Efficacy Update Report (Primary Censoring Rule; ITT Population)

CI = confidence interval; HR = hazard ratio; ITT = intention to treat; NR = not reached; OS = overall survival.

Note: Database cut-off date was June 14, 2021.

^aFrom product-limit (Kaplan-Meier) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

^cOne-sided p value based on log-rank test stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

Source: Clinical Study Report efficacy update report.²⁶

Figure 8: Kaplan-Meier Plot of Overall Survival in the Efficacy Update Report (ITT Population)

+ = censored observations; ITT = intention to treat.

Note: Database cut-off date: June 14, 2021.

Source: Clinical Study Report efficacy update report.²⁶

Table 31: Analysis of Disease-Free Survival at Efficacy Update Report (Primary Censoring rule; ITT Population)

CI = confidence interval; DFS = disease-free survival; HR = hazard ratio; ITT = intention to treat; NR = not reached.

Note: Database cut-off date: June 14, 2021.

^aFrom product-limit (Kaplan-Meier) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

^cOne-sided p value based on log-rank test stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

Source: Clinical Study Report efficacy update report.²⁶

Figure 9: Kaplan-Meier Plot of Disease-Free Survival Based on the Investigator Assessment in the Efficacy Update Report (Primary Censoring Rule; ITT Population)

+ = censored observations; ITT = intention to treat.

Note: Database cut-off date was June 14, 2021.

Source: Clinical Study Report efficacy update report.²⁶

Table 32: Prespecified Subgroup Analysis of DFS at IA1

CI = confidence interval; DFS = disease-free survival; HR = hazard ratio; IA1 = first interim analysis; M0 = patients with no distant metastases; M1 NED = patients who present not only with the primary kidney tumour but also solid, isolated, soft tissue metastases that were completely resected at the time of nephrectomy (synchronous) or ≤ 1 year from nephrectomy (metachronous).

Source: Clinical Study Report.¹²

Figure 10: Post Hoc Subgroup Analysis of DFS According to the Recurrence Risk Subgroups in the Efficacy Update Report — (ITT Population)

DFS = disease-free survival; ITT = intention to treat.

Source: Sponsor submission⁴¹

Table 33: Sensitivity Analysis of Disease-Free Survival at IA1 (Sensitivity Censoring Rule; ITT Population)

CI = confidence interval; DFS = disease-free survival; HR = hazard ratio; IA1 = first interim analysis; NR = not reached.

Note: Database cut-off date: December 14, 2020.

^aFrom product-limit (Kaplan-Meier) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

^cOne-sided P value based on log-rank test stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

Source: Clinical Study Report.¹²

Figure 11: Kaplan-Meier Plot of Disease-Free Survival (Sensitivity Censoring Rule; ITT Population)

+ = censored observations; ITT = intention to treat.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Table 34: Sensitivity Analysis of Disease-Free Survival at IA1 (Sensitivity Analysis with Additional Stratum for Baseline Disease Status Based on BICR Review of Baseline Scan Only; ITT Population)

BICR = blinded independent central review; CI = confidence interval; DFS = disease-free survival; HR = hazard ratio; IA1 = first interim analysis; ITT = intention to treat; NR = not reached.

Notes: Analyses are based on primary censoring rule.

Database cut-off date: December 14, 2020.

^aFrom product-limit (Kaplan-Meier) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by baseline disease status by BICR (NED by BICR vs. Non-NED by BICR), then within NED by BICR further stratified by randomization strata: M0 vs. M1 NED by investigator and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

^cOne-sided P value based on log-rank test stratified by baseline disease status by BICR (NED by BICR vs. Non-NED by BICR), then within NED by BICR further stratified by randomization strata: M0 vs. M1 NED by investigator and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

Source: Clinical Study Report.¹²

Table 35: Sensitivity Analysis of Disease-Free Survival by BICR at IA1 (Sensitivity Analysis in Participants With Baseline Evidence of Disease [Non-NED] Based on BICR of Baseline Scan Only and Censored at Baseline; ITT Population)

BICR = blinded independent central review; CI = confidence interval; DFS = disease-free survival; IA1 = first interim analysis; ITT = intention to treat; HR = hazard ratio; NR = not reached.

Notes: Baseline non-NED was assessed by BICR review of baseline scan only.

Database cut-off date was December 14, 2020.

^aFrom product-limit (Kaplan-Meier) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

^cOne-sided P value based on log-rank test stratified by metastasis status (M0 vs. M1 NED by investigator) and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

Source: Clinical Study Report.¹²

Figure 12: Kaplan-Meier Plot of Disease-Free Survival (Sensitivity Analysis in Participants With Baseline Evidence of Disease [Non-NED] Based on BICR of Baseline Scan Only and Censored at Baseline; ITT Population)

+ = censored observations; BICR = blinded independent central review; ITT = intention to treat.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Figure 13: Kaplan-Meier Plot of Event-Free Survival Based on BICR (Baseline Disease Status Based on BICR Review of Both Baseline and Postbaseline Scans) at IA1 — Full Analysis Set

+ = censored observations; IA1 = first interim analysis.

Notes: For participants who were assessed to have baseline disease based on BICR review of both baseline and postbaseline scans, the date of disease progression is used as event date if it occurs.

Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Table 36: Event-Free Survival at IA1 Based on BICR (Baseline Disease Status Based on BICR Review of Both Baseline and Postbaseline Scans) — Efficacy Analysis Set (ITT Population)

BICR = blinded independent central review; CI = confidence interval; EFS = event-free survival; HR = hazard ratio; IA1 = first interim analysis; ITT = intention to treat; NR = not reached.

Notes: For participants who were assessed to have baseline disease based on BICR review of both baseline and postbaseline scans, the date of disease progression is used as event date if it occurs.

Database cut-off date was December 14, 2020.

^aFrom product-limit (Kaplan-Meier) method for censored data.

^bBased on Cox regression model with Efron’s method of tie handling with treatment as a covariate stratified by baseline disease status by BICR (NED by BICR vs. Non-NED by BICR), then within NED by BICR further stratified by randomization strata: M0 vs. M1 NED by investigator and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

^cOne-sided P value based on log-rank test stratified by baseline disease status by BICR (NED by BICR vs. Non-NED by BICR), then within NED by BICR further stratified by randomization strata: M0 vs. M1 NED by investigator and ECOG PS (0 vs. 1), US participant (Yes vs. No) within M0 group by investigator.

Source: Clinical Study Report.¹²

Table 37: Concordance of Disease Recurrence Assessments at IA1 (ITT Population)

BICR = blinded independent central review; EDR = early discrepancy rate, calculated as BICR disagreed / (BICR agreed + BICR disagreed); IA1 = first interim analysis; ITT = intention to treat; LDR = late discrepancy rate, calculated as BICR declared at earlier time / (BICR declared at earlier time + BICR disagreed).

Notes: Number of participants with disease recurred, not recurred and no postbaseline assessments by investigator are used as the denominators for the percentage calculation for the corresponding block of rows.

Discrepancy on disease recurrence time is counted when more than 1 imaging assessment time point difference is observed.

A difference in EDR ≤ −0.05 or a difference in LDR >  = 0.075 are suggestive of a systematic bias in the investigator assessment favouring the pembrolizumab arm.²⁸

Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Table 38: Summary of Pathological Assessments for Disease Recurrence at IA1 (ITT Population)

BICR = blinded independent central review; IA1 = first interim analysis; ITT = intention to treat; INV = investigator assessment.

Notes: Pathological assessment of radiographic disease recurrence by investigator’s review is optional. Disease recurrence by BICR is determined by BICR of imaging alone, regardless of pathological assessment results.

Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Table 39: Analysis of Change From Baseline to Week 52 in PROs Defined as Exploratory End Points in KEYNOTE-564 Trial (EORTC QLQ-C30) — PRO FAS Population (IA1)

CFB = change from baseline; CI = confidence interval; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FAS = full analysis set; IA1 = first interim analysis; LSM = least squares mean; PRO = patient-reported outcome; SD = standard deviation; VAS = visual analogue scale.

Notes: For baseline and week 52, N is the number of patients in each treatment group with non-missing assessments at the specific time point; for change from baseline, N is the number of patients in the analysis population in each treatment group.

Database cut-off date: December 14, 2020.

^aBased on a cLDA model with the PRO scores as the response variable with covariates for treatment by study visit interaction, stratification factors metastasis status (M0 vs. M1 NED), and within M0 group further stratified by ECOG PS (0 vs. 1) and US participant (Yes vs. No) as covariates.

Source: Clinical Study Report.¹²

Figure 14: LS Mean Change From Baseline to Week 52 and 95% CI in EORTC QLQ-C30 Global Health Status/QoL and Functional Scales (PRO FAS Population)

CI = confidence interval; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FAS = full analysis set; LS = least squares; PRO = patient-reported outcome; QoL = quality of life.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Figure 15: LS Mean Change From Baseline to Week 52 and 95% CI in EORTC QLQ-C30 Symptom Scales (PRO FAS Population)

CI = confidence interval; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FAS = full analysis set; LS = least squares; PRO = patient-reported outcome; QoL = quality of life.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Table 40: Definition of Deterioration, Stability, Improvement, Stability Plus Improvement for the Analyses of PROs — EORTC QLQ-C30 Global Health Status/QoL, and Functional (Physical Functioning) Scale and FKSI-DRS

EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FKSI-DRS = Functional Assessment of Cancer Therapy Kidney Symptom Index - Disease-Related Symptoms; PRO = patient-reported outcome; QoL = quality of life.

Source: Clinical Study Report.¹²

Table 41: Overall Improvement and Stability Rate for PROs at IA1 (FKSI-DRS Score, EORTC QLQ-C30 Global Health Status, EORTC QLQ-C30 Physical Functioning) — PRO FAS Population

CI = confidence interval; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire Core 30; FKSI-DRS = Functional Assessment of Cancer Therapy Kidney Symptom Index–Disease-Related Symptoms; IA1 = first interim analysis; NA = not applicable.

Note: Database cut-off date: December 14, 2020.

^aBased on binomial exact confidence interval method.

^bBased on Miettinen and Nurminen method with population-based weighting stratified by metastasis status (M0 vs. M1 NED), and within M0 group further stratified by ECOG PS (0 vs. 1) and US participant (Yes vs. No).

^cTwo-sided P value for testing. H0: difference in % = 0 vs. H1: difference in % does not equal 0.

Source: Clinical Study Report.¹²

Figure 16: Between-Treatment Comparisons in Adverse Events (≥ 10% Incidence) at IA1 (APaT Population)

APaT = All Participants as Treated; IA1 = first interim analysis.

Note: Database cut-off date: December 14, 2020.

Source: Clinical Study Report.¹²

Appendix 4: Description and Appraisal of Outcome Measures

Note this appendix has not been copy-edited.

Aim

To describe the following outcome measures and review their measurement properties including validity, reliability, responsiveness to change, and minimally important difference:

• European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire – Core 30 (EORTC QLQ-C30)

• Functional Assessment of Cancer Therapy Kidney Cancer Symptom Index – Disease-Related Symptoms (FKSI-DRS)

• 5-Level EQ-5D 5 (EQ-5D-5L)

Findings

Table 42: Summary of Outcome Measures and Their Measurement Properties

ECOG PSR = Eastern Cooperative Oncology Group Performance Status rating; EORTC QLQ-C30 = European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire – Core 30; EQ-5D-5L = 5-Level EQ-5D; FACT-G = Functional Assessment of Cancer Therapy-General; FKSI-DRS = Functional Assessment of Cancer Therapy Kidney Cancer Symptom Index – Disease-Related Symptoms; GRCS = Global Rating of Change Scale; HRQoL = health-related quality of life; MID = minimally important difference; RCC = renal cell carcinoma; VAS = visual analogue scale.

European Organisation for the Research and Treatment of Cancer Quality of Life Questionnaire – Core 30 (EORTC QLQ-C30)

The EORTC QLQ-C30 is 1 of the most commonly used instruments in oncology clinical trials.⁴² The questionnaire is a cancer-specific, multidimensional, patient-reported outcome measure used to assess change in HRQoL in response to treatment.¹⁸ The core questionnaire of the EORTC QLQ-C30 consists of 30 items (questions) that are scored to create 5 multi-item functional scales (physical, role, cognitive, emotional, and social), 3 multi-item symptom scales (fatigue, pain, nausea and vomiting), 6 single-item symptom scales (dyspnea, insomnia, appetite loss, constipation, diarrhea, and financial impact [these are symptoms commonly reported by patients with cancer]), and a global health status/quality of life scale.¹⁹ The recall period is 1 week.¹⁹ Most questions have 4 response options (“not at all,” “a little,” “quite a bit,” “very much”), with scores on these items ranging from 1 to 4. For the 2 items that form the global health status/quality of life scale, the response format is a 7-point Likert-type scale, with anchors of 1 (very poor) to 7 (excellent).¹⁹ Raw scores for each scale are computed as the average of the items that contribute to a particular scale.¹⁹ Each raw scale score is converted to a standardized score that ranges from 0 to 100. A higher score on the functional scale, the symptom scale and items, and the global health status/quality of life scale indicates a higher level of functioning, a higher level of symptomatology, and a higher level of quality of life, respectively.¹⁹