3.2.3. Brain Metastases

About 10% of non-oncogene addicted patients with NSCLC have brain metastases at diagnosis and 25–40% develop brain metastasis during the course of the disease. A pooled analysis of the CheckMate 017 and CheckMate 057 trials showed that 11% of the included patients had brain metastasis at baseline, but no detailed information about the intracranial efficacy of nivolumab were provided in the primary publications [4]. However, nivolumab was more effective than docetaxel in terms of OS in the entire analyzed patient population [4]. CheckMate 012 (NCT01454102) was a phase I, multicohort study evaluating nivolumab alone or in combination with other therapies for the treatment of patients with advanced NSCLC and untreated brain metastases (12 patients, arm M) [18]. Intracranial response was evaluated with magnetic resonance imaging [18]. The ORR was 16.7% (two patients) in that small study group; however, progressive disease was observed in the majority of patients [18].

Nivolumab therapy is routinely used in patients who have undergone primary resection or irradiation for brain metastases and whose clinical condition improved after receiving local treatment. Retrospective analyses of real-world data showed that nivolumab has intracranial activity [19–21]. Twenty-six percent of patients with non-squamous NSCLC in the Italian EAP had asymptomatic or controlled brain metastases [19]. The disease control rate was 40% and the ORR was 17%. The median OS in patients with asymptomatic or controlled brain metastases was 8.6 months (95% CI 6.4–10.8) compared with 11.3 months (95% CI 10.2–12.4) for the entire cohort [20]. In the cohort with squamous NSCLC, 10% of 372 patients had asymptomatic brain metastases. The median OS was 5.8 months (95% CI 1.9–9.8) [21]. A direct comparison of the efficacy of nivolumab in patients with and without brain metastases showed significant differences in OS [14]. In a group of 188 patients, 22% had brain metastases. The median OS was 5.09 months (95% CI 0.3–9.8) in the patients with brain metastases versus 14.8 months (95% CI 11.5–17.3) in patients without brain metastases [14]. In another cohort, in which 14.8% of 472 patients had brain metastases, the median OS reached 9 months (95% CI 5.5–13.3) in patients with brain metastases and 13.1 months (95% CI 11.5–17.1; *p* = 0.007) in patients without brain metastases [12]. Some studies have identified brain metastases as an independent negative prognostic factor [8], but others have not [6,7,9,10].

#### 3.2.4. Elderly Patients

More than 40% of patients in the CheckMate 017 and CheckMate 057 populations were over 65 years of age, including about 7% of patients who were over 75 years of age [2,3]. Nivolumab was effective in the whole group, although for patients over 75 years of age the clinical benefit was uncertain (HR 0.9; 95% CI 0.43–1.87) [3]. The findings of the phase II CheckMate 171 trial have been published recently [22]. Overall, 811 patients with previously treated advanced squamous NSCLC were included, of whom 278 were aged over 70 years and 125 were aged over 75 years [22]. The median OS was similar in all age groups: 10.0 months (95% CI 9.2–11.2) in all patients, 10.0 months (95% CI 8.3–11.4) in those aged over 70 years, and 11.2 months (95% CI 7.9–14.2) in those aged over 75 years. The safety profile was similar across age-determined populations; however, low-grade diarrhea was more common in patients over 70 years of age than in those aged 70 or younger [22]. AEs were reported

in 13.9% of all patients, in 15.8% of those aged over 70 years, and in 18.4% of those aged over 75 years [22]. In an Italian population of 371 patients with squamous NSCLC, OS was reduced in patients aged 75 years or older (5.8 months, 95% CI 3.5–8.1) versus patients aged under 65 years (8.6 months, 95% CI 5.2–11.9), patients aged 65 to less than 75 years (8.0 months, 95% CI 5.6–10.4), and the overall population (7.9 months, 95% CI 6.2–9.6) [23]. Discontinuation rates due to treatment-related AEs were low irrespective of age (4–5%) [24]. However, a retrospective analysis of 324 Belgian patients with NSCLC showed no significant difference between older (≥70) and younger (<70 years) patients in terms of PFS (4 months versus 3.7 months, *p* = 0.483) and OS (9.3 months versus 8.4 months, *p* = 0.638) [25]. The incidence of AEs of all grades and of grade 3–4 AEs was also similar between age groups [25]. Similarly, in a group of Italian patients with non-squamous NSCLC, 522 of 1588 patients were over 70 years of age; these patients reached a median OS of 11.5 months (95% CI 10.0–13.0), while for the 232 patients aged over 75 years OS was 12.0 months (95% CI 9.2–14.8) [6]. There were no significant differences in the incidence of treatment-related AEs in the subgroups defined by age (6–7% of AEs were grade 3–4) [6]. Some studies have confirmed that treatment outcomes in clinical practice are not affected by age [11,12,26,27], whereas others have reported nivolumab treatment to have less favorable results in patients aged over 75 years [9].
