Low‐dose pembrolizumab in the treatment of advanced non‐small cell lung cancer

Abstract A dose of 200 mg 3‐weekly of pembrolizumab was approved by the Food and Drug Administration (FDA) as treatment for advanced non‐small cell lung cancer (NSCLC) without oncogenic drivers. This is despite evidence showing no difference in efficacy with 2 mg/kg. Our study aimed to assess the efficacy of a lower fixed dose of 100 mg, which is closer to 2 mg/kg weight‐based dose in an average‐sized Asian patient. All patients receiving pembrolizumab for advanced NSCLC from January 2016 to March 2020 in National University Hospital, Singapore, were included in this retrospective observational study. The effect of pembrolizumab 100 mg (Pem100) vs 200 mg (Pem200) upon survival outcomes, toxicity and cost were examined. One hundred fourteen patients received pembrolizumab. Sixty‐five (57%) and 49 (43%) received Pem100 and Pem200, respectively. There was no difference in progression‐free survival (PFS) and overall survival (OS) between Pem100 vs Pem200 as a single agent (PFS: 6.8 vs 4.2 months, hazard ratio [HR] 0.72, 95% confidence interval [CI] 0.36‐1.46, P = .36; 9 month OS: 58% vs 63%, HR 1.08, 95% CI 0.48‐2.41, P = .86) and when combined with chemotherapy (9‐month PFS: 60% vs 50%, HR0.84, 95% CI 0.34‐2.08, P = .71; 9‐month OS: 85% vs 58%, HR 0.27, 95% CI 0.062‐1.20, P = .09). No significant difference in response rate or ≥G3 immune‐related toxicities between Pem100 and Pem200 was observed. A cost minimisation analysis evaluating the degree of cost savings related to drug costs estimated a within study cost saving of SGD4,290,912 and cost saving per patient of SGD39,942 in the Pem100 group. A 100 mg of pembrolizumab appears to be effective with reduction in cost. A randomised trial should be done to investigate a lower dose of pembrolizumab.


| INTRODUCTION
Worldwide, lung cancer is the most common cancer and leading cause of cancer death. 1 Cancer immunology has enabled the development of immune modulators that have markedly altered the treatment landscape for patients with advanced non-small cell lung cancers (NSCLC) that do not harbour oncogenic drivers. 2 Pembrolizumab is a fully humanised immunoglobulin G4 monoclonal antibody directed against the programmed cell death protein 1 (PD-1) receptor, antagonising the interaction between itself and its ligand, resulting in anti-tumour immune response. 3,4,5 Since the Food and Drug Administration (FDA) approval of pembrolizumab in 2015 6 in pre-treated NSCLC, 7,8,9,10 pembrolizumab has rapidly transited to standard of care first line treatment as a monotherapy 11,12,13 or with chemotherapy 14,15,16,17 for NSCLC with no oncogenic drivers.
In the landmark KEYNOTE-001, an ex vivo pharmacokinetics study of PD-1 receptor saturation found complete peripheral target engagement at 1 mg/kg. In the initial dose escalation cohort, durable anti-tumour activity across all patient cohorts from 1-10 mg/kg once every 3 weeks was observed. 18 The subsequent expansion cohorts demonstrated that the efficacy of 2 mg/kg 3-weekly was similar to higher dose 10 mg/kg 2-weekly regimen but a dose of <2 mg/kg was not examined. 7,9,12,19 Despite the pharmacokinetics profile showing PD-1 receptor saturation at 1 mg/kg and clinical efficacy at 2 mg/kg, a flat dose of 200 mg 3-weekly was used in Phase III trials leading to the FDA approval of 200 mg every 3 weeks. 14,15,16,17 More recently, the FDA granted accelerated approval of a new dosing regimen of 400 mg every 6 weeks. 6 Goldstein et al performed an economic analysis comparing the FDA approved fixed dosing and personalised dosing at 2 mg/kg and estimated a cost saving of USD$0.8 billion annually to the United States healthcare system based on an average weight of 75 kg in an American adult patient. 20,21,22,23 The average weight of an Asian patient is 60 kg giving an average personalised dose of 120 mg in this population. 24 Given the lack of benefit demonstrated by pembrolizumab at doses above 2 mg/kg, the lower weight of Asian patients, economic benefits of a lower dose and packaging of pembrolizumab in 100 mg vials, a fixed dose of 100 mg pembrolizumab required evaluation in an Asian population.
Using a retrospective observational design, our study aimed to evaluate the efficacy of low-dose pembrolizumab (Pem100) compared with standard-dose pembrolizumab (Pem200) in the treatment of NSCLC.

| Patients and treatment
All patients receiving palliative intent pembrolizumab for advanced NSCLC with or without chemotherapy between January 2016 and March 2020 in an academic tertiary medical centre (National University Hospital, Singapore) were identified retrospectively from the pharmaceutical database. Baseline patient demographics, tumour and treatment characteristics were extracted from the electronic medical records.
The dose of 100 mg was routinely delivered based on an approximate 2 mg/kg weight-based dose, for patients who did not have an adequate financial reimbursement plan or based on physician's preference. Local protocols continue treatment until disease progression, unacceptable toxicities, death, patient's decision to stop treatment or after a total of 35 cycles of pembrolizumab although some patients who remained progression free after 35 cycles continued treatment.

| Response evaluation
Chest and/or abdominal CT scans were performed by clinicians every 8-12 weeks as part of routine clinical care, to evaluate patient's response and assess for disease progression. The scans were evaluated by investigators retrospectively. In line with the KEYNOTE studies efficacy analysis was examined only in patients without an oncogenic driver mutation. A systemic response to pembrolizumab was measured by standard Response Evaluation Criteria in Solid Tumours (V.1.1). 25 The best response was classified as progressive disease (PD), stable disease (SD), partial response (PR) and complete response. Progression-free survival (PFS) was measured from time of initiation of drug to disease progression by RECIST or death due to any cause. Overall survival (OS) was measured from time of initiation of drug to death. Safety analysis examined the incidence of ≥ Grade 3 immune-related adverse events (irAEs) and adverse events (AEs) as recorded by clinicians.

| IHC of PDL1
The PD-L1 IHC 22C3 pharmDx assay 26,27 was used to assess PD-L1 expression in formalin-fixed tumour samples obtained at the What's new?
Pembrolizumab, a monoclonal antibody directed against the PD-1 receptor, has received FDA approval for the treatment of lung cancer at a fixed dose of 200 mg every 3 weeks. However, doses above 2 mg/kg show a lack of benefit, calling for further evaluation in Asian populations. This retrospective observational study demonstrates the efficacy of a lower fixed dose of pembrolizumab (100 mg every 3 weeks) compared with standard-dose pembrolizumab. The results also confirm the clinical activity of pembrolizumab at a lower dose than 2 mg/kg every 3 weeks, which could provide considerable cost savings to patients and the health system. time of diagnosis of metastatic disease. PD-L1 clone 22C3 is from Dako and stained on Roche Ventana Benchmark Ultra ISH/IHC autostainer with Optiview Polymer Detection Kit. Interpretation of PD-L1 expression is based on the interpretation guide provided by Dako and was characterised according to tumour proportion score (TPS). 28

| Statistical and economic analysis
Differences in the baseline characteristics of patients receiving Pem100 and Pem200 were evaluated using the Fisher's exact test.
Survival analyses were performed using the Kaplan-Meier method and were compared using a log-rank test. Multi-variable Cox proportional hazard regression models were used to assess the relationship between baseline factors (including treatment) and survival. A P-value of <.05 was considered statistically significant. All statistical tests were two-sided and were performed using IBM SPSS Statistics Version 22.
Based on an acceptance of non-inferior survival and toxicity outcomes, a limited economic evaluation was carried out using a cost-minimisation approach. 29 This assessed the monetary savings available from the use of Pem100 instead of Pem200 based on the total and median cycles of pembrolizumab received by the study population and the price of a 100 mg vial of pembrolizumab.
Sensitivity analysis considered the potential savings within the study population if all patients and if patients weighing ≤100 kg (translating to a dose of at least 1 mg/kg) were to receive Pem100. Given the identical regimens and observed clinical outcomes, all other costs were assumed to remain constant.

| Survival
Ten patients with oncogenic driven NSCLC who received pembrolizumab were excluded from the analysis. Median duration of follow-up was 14.8 months.
The median PFS of Pem100 vs Pem200 as a single agent was not statistically significant at 6.8 vs 4.2 months (HR 0.60, 95% CI 0.30-1.22, P = .16) ( Figure 1A ). PFS for Pem100 and Pem200 did not differ in all subgroups examined ( Figure 2

| Response rates
We analysed the response rates (RRs) of 88 patients who received pembrolizumab in the first line setting.

| Toxicities
Eighteen patients discontinued treatment due to toxicities. There was no dose relationship between pembrolizumab and serious irAEs. The rates of G3 or more irAEs between Pem100 and Pem200 were observed to be 17% vs 22%, P = .5. further sensitivity analysis of Pem100 in patients weighing ≤100 kg demonstrates a total cost savings of SGD 70065706.

| DISCUSSION
To our knowledge, our study represents the largest cohort to date in which the efficacy of a lower fixed dose of 100 mg pembrolizumab given 3-weekly is demonstrated. 30 We also confirm the clinical activity of pembrolizumab at a dose lower than 2 mg/kg 3-weekly. Finally, given no difference was identified in the clinical outcomes of the two regimens a cost minimisation analysis was used to examine the cost saving provided by Pem100. This was not planned a priori and simply provides an indication of possible savings. The costs assessed are only those of the drug and do not include regimen related costs such as drug administration, premedication, clinic visits, subsequent therapy and AE management.
Based on the study outcomes these costs are not anticipated to vary; however, further formal assessment of the cost-utility of Pem100 should be considered alongside any future randomised study.
Despite these limitations, our study is the first to suggest clinical efficacy of pembrolizumab at a fixed dose of 100 mg 3-weekly. This lower dose could be efficacious and provide considerable cost savings to both patients and the health system more widely. Such savings could be redistributed to other health needs. [41][42][43] 5 | CONCLUSION In our study, pembrolizumab had efficacy at a dose of 100 mg 3-weekly. With the expanding role of immune checkpoint inhibitors in many tumour types, the principles and solutions discussed here will be highly relevant to oncologists, policymakers and patients alike. A randomised prospective trial is now required to further investigate the role and cost-effectiveness of lower-fixed dosing of pembrolizumab at 100 mg 3-weekly or 200 mg 6-weekly.

CONFLICT OF INTEREST
A. Jeyasekharan has received consultancy fees from Turbine Ltd,

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author.

ETHICS STATEMENT
The study was approved by the National Health Group Domain Specific Review Board (NHG DSRB) (Reference number: 2017/012654) and was conducted in accordance to the Declaration of Helsinki provision. Informed consent was waived because of the retrospective design of the study.