Do Elderly Lung Cancer Patients Aged ≥75 Years Benefit from Immune Checkpoint Inhibitors?

Lung cancer patients ≥75 years represent nearly 40% of all lung cancer patients and continue to increase. If elderly patients have a good performance status and adequate organ function, they can be treated the same as non-elderly patients. However, few comparative studies limited to elderly patients (≥75 years) have been conducted. We review the evidence on using immune check inhibitors for the treatment of elderly patients (≥75 years old) with advanced non-small cell lung cancer. Prospective randomized or non-randomized, retrospective, registrational, insurance-based, and community-based studies have shown that elderly (≥75 years) and non-elderly patients are similarly treated with immune check inhibitors effectively and safely. However, such analyses have not shown that immune check inhibitors are significantly more effective than chemotherapy alone. In addition, patient selection might be critically performed to administer immune check inhibitors in the elderly because they are more likely to have a poor performance status with comorbidities, which lead to little benefit, even in non-elderly patients. There is a need for more evidence showing the benefit of immune check inhibitors in non-small cell lung cancer patients ≥75 years.


Introduction
Before the era of molecular-targeted agents and immune checkpoint inhibitors (ICIs) such as a programmed cell death protein 1 (PD-1) antibody, the age of eligibility of elderly patients with advanced non-small cell lung cancer (NSCLC) in phase III randomized trials was often defined as 65 or 70 years [1], and only 30-50% of patients enrolled in these trials were aged ≥75 years [1]. In addition, there have been few recent studies on molecular-targeted agents and ICIs in patients ≥75 years of age [2].
The median age at diagnosis of lung cancer was 71 years old and patients ≥75 years old was 36.3% according to Surveillance, Epidemiology, and End Results (SEER) Cancer Statistics Review in 2013-2017 [3]. In addition, the median age at nivolumab or pembrolizumab initiation in real-world metastatic NSCLC patients was 69 years; 27% were aged 75 or older [4]. The Japanese guidelines for the diagnosis and treatment of lung cancer define elderly lung cancer patients as ≥75 years [5]. In 2015, the lung cancer incidence was highest in males aged 70-74 years, followed by males aged 75-79 years, and in females aged ≥ 85 years, followed by females aged 70-74 years [6]. Therefore, a cutoff age of 75 years is considered reasonable for distinguishing elderly from non-elderly groups among lung cancer patients. Ferrara et al. reviewed the relationship of immunotherapy resistance with immunosenescence, which was probably linked to the continuous and progressive remodeling of immune functions with ageing [7]. The loss of senescence markers such as CD27 and CD28 or the expression of Tim-3 and CD57 on T cells was associated with resistance to ICIs [8]. Immunosenescence may be related to a decreased response to ICIs or to an increased risk of adverse events, especially in elderly NSCLC patients (≥75 years) [7].
If elderly patients have a good performance status (PS) and adequate organ function, they can be treated the same as non-elderly patients. However, few comparative studies limited to elderly patients (≥75 years) have been conducted. Even in a recent review of elderly NSCLC patients, the cut-off age was 65 or 70 years [9]. Although there was no upper age limit in the eligibility criteria for most recent prospective studies, the extent to which the results derived from non-elderly patients can be extrapolated to elderly patients is unknown. Here, we review the evidence on using ICIs for the treatment of elderly patients (≥75 years old) with advanced lung cancer, especially NSCLC.

Standard Anticancer Drug Therapies for Elderly Patients with Advanced NSCLC
Patients with advanced NSCLC with abnormalities in the epidermal growth factor receptor, anaplastic lymphoma kinase fusion, B-Raf proto-oncogene, ROS proto-oncogene 1 fusion, or neurotrophic tropomyosin receptor kinase fusion gene are recommended treatment with appropriate molecular-targeted drugs, which have also been used in the elderly [9]. In addition to these markers, it is important to evaluate the immunohistochemical expression of programmed death ligand 1 (PD-L1), as PD-L1 expression is a confirmed predictive factor for the treatment outcome of ICI monotherapy in NSCLC patients. Until several years ago, a single third-generation cytotoxic anticancer drug (docetaxel, vinorelbine, or gemcitabine) was recommended for NSCLC [9]. Recently, the results of a phase 3 study of carboplatin plus pemetrexed followed by maintenance pemetrexed compared with docetaxel monotherapy in non-squamous NSCLC patients ≥75 years old was reported [10]. Non-inferiority in survival time and superiority in progression-free survival were found in the drug combination group, and the regimen has become a standard treatment.
There is no randomized study comparing ICIs with chemotherapy in elderly NSCLC patients. If PS is 0 or 1 without severe comorbidities, pembrolizumab monotherapy may be used for PD-L1-positive cases [11,12]. In addition, regardless of PD-L1 expression, platinum combination therapy plus ICI is the standard treatment for non-elderly patients with a PS of 0 to 1 [13][14][15][16], but its usefulness in the elderly is not clear. There may be additional value in using ICI for carboplatin plus pemetrexed or carboplatin plus (nab-) paclitaxel combinations in this population. Later, we will review the use of ICI with or without chemotherapy for the treatment of advanced NSCLC patients ≥75 years old.

Standard Second-Line Treatment for Elderly Patients with NSCLC
There have been few randomized studies on second-line treatments for elderly patients (≥75 years old) [17,18]. When pembrolizumab is administered as a first-line treatment for PD-L1-positive NSCLC, a cytotoxic anticancer drug is used as a second-line treatment. In patients who do not receive pembrolizumab in first-line therapy, ICI such as pembrolizumab, nivolumab, or atezolizumab are candidates for second-line therapy. PD-L1 positivity (≥1%) is an essential criterion for the use of pembrolizumab, but nivolumab and atezolizumab can be used irrespective of PD-L1 expression [19].

Initial Treatment for Extensive-Stage Small-Cell Lung Cancer
The IMpower133 and CASPIAN phase 3 studies showed the efficacy of adding ICIs (atezolizumab or durvalumab) to first-line treatments comprising platinum and etoposide in patients with extensive-stage small cell lung cancer [20,21]. In the IMpower133 study, only 19 patients (9.5%) were ≥75 years of age in the carboplatin-etoposide-atezolizumab group and 22 (10.9%) in the carboplatin-etoposide-placebo group. In the CASPIAN study, the median age (interquartile range) was 62 (58-68) years in the durvalumab-etoposide plus cisplatin or carboplatin arm and 63 years (57-68) in the etoposide plus cisplatin or carboplatin arm. KEYNOTE-604 study comparing pembrolizumab plus etoposide and platinum with placebo plus etoposide and platinum for patients with previously untreated extensive-stage small cell lung cancer was also recently reported. Median age (range) was 64  in the former arm and 65  in the latter arm [22]. The numbers of patients ≥75 years old in both studies were not stated; thus, the efficacy and safety of ICIs in patients ≥75 years old with extensive-stage small cell lung cancer remain unknown.

Efficacy of ICIs in Elderly NSCLC Patients (≥75 years) in Phase 3 Studies
The hazard ratios (HRs) for overall survival (OS) in patients <75 and ≥75 years of age with advanced NSCLC in phase III studies are shown in Table 1.
In the KEYNOTE-024 and KEYNOTE-042 studies, which compared pembrolizumab monotherapy with chemotherapy in the first-line setting, a sub-analysis of patients ≥75 years old was not initially reported [11,12]; however, recently, a pooled analysis of patients ≥75 years old in those trials was published [23]. The HR (95% confidence interval [CI]) for OS in the KEYNOTE-024 study was 0.64 (0.42-0.97) in patients <75 years and 0.49 (0.17-1.39) in those ≥75 years. The HR for OS in the KEYNOTE-042 trial was 0.79 (0.68-0.92) in patients <75 years and 0.89 (0.59-1.35) in those ≥75 years. No significant difference was observed in the sub-analysis, but the favorable trend observed in the pembrolizumab group was maintained.
In the CheckMate 017 and CheckMate 057 studies, nivolumab was found to prolong OS compared with docetaxel as salvage therapy in patients with squamous cell carcinoma or non-squamous cell carcinoma [24,25]. CheckMate 017 showed significant differences in HRs (95% CI) of 0.52 (0.35-0.75) and 0.56 (0.34-0.91) in patients < 65 years and 65-74 years, respectively. The HR (95% CI) in patients ≥75 years was 1.85 (0.76-4.51), indicating a favorable therapeutic effect in the docetaxel group, but this effect was considered to be skewed by the small number of cases (29 patients). The HR in patients ≥75 years in the CheckMate 057 study was 0.90 (0.43-1.87), indicating similar effectiveness in both arms, although the sample size was small (43 patients). CheckMate 227 was a phase 3 study comparing nivolumab plus ipilimumab with chemotherapy for first-line treatment of advanced NSCLC [26]. An OS benefit was observed irrespective of PD-L1 expression level. The HRs (95% CI) were 0.70 (0.58-0.85), 0.76 (0.61-0.95), and 0.84 (0.55-1.29) in patients <65 years, ≥65 to <75, and ≥75 years old, respectively. The efficacy of combined ICIs in elderly patients (≥75 years) remains unknown.

Efficacy of ICIs for Elderly NSCLC Patients in Non-Randomized Studies
Non-randomized studies of elderly NSCLC patients treated with ICI monotherapy are summarized in Table 2.  CheckMate 153 [33] prospectively examined the safety and efficacy of nivolumab in patients with advanced NSCLC including patients aged ≥70 years with a poor PS, who are typically under-represented or excluded from randomized studies. The mOS was comparable in the overall population (9.1 months, n = 1426) and in patients ≥70 years (10.3 months, n = 556). CheckMate 171 [34], which had a similar design to CheckMate 153, only included patients with squamous NSCLC. The OS and response rate (RR) corresponding to the entire population (n = 811), patients ≥70 years (n = 278), and those ≥75 years (n = 125) were 10.0 months and 11.0%, 10.0 months and 12.6%, and 11.2 months and 13.6%, respectively. Thus, nivolumab for advanced, relapsed squamous NSCLC seems to be similarly effective between elderly and non-elderly patients.
Yamaguchi et al. [40] retrospectively evaluated the efficacy of subsequent-line nivolumab or pembrolizumab monotherapy in elderly Japanese patients (aged ≥75 years) with a median age of 77 years (range, 75-87). The mOS was 16.0 months (95% CI: 12.1-19.8) in all patients (n = 131). There was no significant difference in the mOS between patients 75-79 years old (13.1 months) and those ≥80 years old (18.9 months) (HR: 1.44; 95% CI: 0.83-2.68; p = 0.2). Thus, ICIs seemed to be as effective in elderly NSCLC patients ≥75 years old as in those <75 years old in prospective or retrospective non-randomized studies in EAP, SEER, and registry cohorts.

Comparison of Adverse Events Between Elderly and Non-Elderly Patients Treated with ICI Monotherapy
ICIs such as nivolumab and pembrolizumab are usually associated with fewer treatment-related adverse events (TRAEs) compared with chemotherapy in randomized studies [11,12,24,25,27]. However, few studies have compared TRAEs in NSCLC patients aged ≥75 or <75 years treated with ICIs. A pooled analysis of KEYNOTE-010, KEYNOTE-024, and KEYNOTE-042 studies showed that TRAEs seemed to slightly increase in patients ≥75 years compared with those aged <75 years, although statistical analysis was not performed [23]. Any-grade TRAEs were observed in 102 patients (68.5%), and grade ≥3 TRAEs were observed in 36 (24.2%) of 149 patients ≥75 years old. Meanwhile, any-grade TRAEs were also observed in 862 (65.2%) and grade ≥3 TRAEs in 224 (16.9%) of 1323 patients <75 years old.

Conclusions and Future Directions
The molecular diagnosis of advanced NSCLC is considered essential for both elderly and non-elderly patients. Several prospective randomized or non-randomized, retrospective, registrational, insurance-based, and community-based studies have shown that elderly (≥75 years) and non-elderly patients are similarly treated with ICIs effectively and safely. However, patient selection is needed to administer ICIs in the elderly because they are more likely to have a poor PS with comorbidities, which lead to little benefit, even in non-elderly patients [41,42]. In addition, sub-analysis of patients ≥75 years has not shown that ICIs are significantly more effective than chemotherapy alone. A prospective randomized study comparing ICIs with chemotherapy or ICIs plus chemotherapy with chemotherapy alone in advanced NSCLC patients ≥75 years might be needed. In addition, a cost-benefit analysis should be performed because ICIs are more expensive than chemotherapy [43,44]. If molecular biomarkers determining the efficacy of ICIs are established, it would be reasonable to use ICIs similarly molecular-targeted agents. There is a need for more evidence showing the benefit of ICIs in NSCLC patients ≥75 years, who represent nearly 50% of all NSCLC patients and will continue to increase in number. At this point, we could not decide if elderly lung cancer patients aged ≥75 years benefit from ICIs.
Funding: This research received no external funding.