3.2.5. *EGFR* Status

Of the patients in the CheckMate 057 trial, 15% had an *EGFR* mutation. Nivolumab was not better than docetaxel in that subset of patients (HR 1.18, 95% CI 0.69–2.0) [3].

In an Italian cohort of patients with non-squamous NSCLC, 102 patients (6.4%) had an *EGFR* mutation [28]. No statistically significant difference in OS was observed in patients with an *EGFR* mutation versus that in those without. OS reached 11 months in patients with *EGFR* wild-type tumors versus 8.3 months in patients with *EGFR*-mutant tumors (HR 1.11, 95% CI 0.84–1.47; *p* = 0.4) [28]. A study by the Galician Lung Cancer Group showed that OS was higher in patients without an *EGFR* mutation than in patients with *EGFR*-mutated NSCLC (12.8 versus 4.8 months, *p* = 0.12) [14]. Although univariate (HR 1.11, 95% CI 0.84–1.45; *p* = 0.46) and multivariate (HR 1.13, 95% CI 0.82–1.56; *p* = 0.45) analysis of 901 Japanese patients (12.9% with *EGFR* mutation) failed to determine the prognostic value of *EGFR* mutation [8], most reports are in line with the CheckMate 057 results and confirmed the negative prognostic value of *EGFR* mutation in patients treated with nivolumab. OS in 25 Canadian patients with *EGFR* mutation was 3.38 months, while in 229 patients with wild-type *EGFR* it was 13.37 months (HR 2.32, 95% CI 1.37–3.93; *p* = 0.002) [12]. A multivariate analysis of 613 patients (15% of whom had an *EGFR* mutation) showed *EGFR* mutation or *ALK* translocation to have negative prognostic value in terms of PFS (HR 1.45, 95% CI 1.12–1.86) [26].

## 3.2.6. Sensitivity to Previous Chemotherapy

A post hoc exploratory multivariate analysis of the CheckMate 057 population suggested that some patients might be at a higher risk of death within the first 3 months of treatment. The following known negative prognostic factors were considered: less than 3 months since last treatment, progressive disease as best response to prior treatment, and an ECOG score of 1 [29].

Real-life experience with nivolumab has shown that sensitivity to previous chemotherapy could have prognostic value. The Netherlands Cancer Institute published the results of 248 patients treated with nivolumab [10]. Of the 189 patients who had a documented response to prior platinum-based doublet therapy, 38.6% had progressive disease as the best response. OS was 13.1 months in patients who had been sensitive to the chemotherapy and only 5.0 months in chemotherapy-refractory patients (HR 1.7, 95% CI 1.108–2.642; *p* = 0.015). An analysis of 221 patients showed that time to progression could also have prognostic value [11]. Patients who had disease progression within 6 months of platinum therapy did worse than those who had a longer PFS than 6 months on platinum therapy (3.7 months versus 11.8 months; HR 0.39, 95% CI 0.26–0.6; *p* < 0.0001) [11]. The positive prognostic

value of both ORR and PFS of more than 6 months since the beginning of prior chemotherapy was presented in another publication [30].

#### *3.3. Safety*

In the CheckMate 057 study, the frequency of any-grade AEs related to nivolumab treatment was 69%, while the frequency of grade 3–4 AEs was 10% [3]. The most common any-grade AEs were fatigue (16%), nausea (12%), decreased appetite (10%), and asthenia (10%). The rate of discontinuation due to nivolumab-related AEs was 5% [3]. In the CheckMate 017 study, any-grade AEs were reported in 58% of patients, while grade 3–4 AEs were reported in 7% of patients, and treatment was discontinued due to nivolumab-related AEs in 3% of patients [2]. The safety profile established in clinical practice seems to be consistent with that determined in clinical trials. The relevant data are summarized in Table 4. The differences in the frequency of any-grade AEs between some publications could be associated with less precise reporting of AEs outside clinical trials.

**Table 4.** Incidence of adverse events (AEs) in patients treated with nivolumab—real-world data.


nd—no data.

### **4. Summary**

Based on the results of the CheckMate 017 and CheckMate 057 studies, nivolumab therapy has become a new standard of care for both squamous and non-squamous NSCLC. Due to the specific inclusion criteria of the clinical trials, the efficacy and safety of nivolumab in real-world practice were not certain. However, some data from the EAP and from post-registration studies have recently been published, which allows for further evaluation.

In general, the real-world results are consistent with those obtained in clinical trials. From a practical point of view, the important question is how to select a subgroup of patients in whom clinical benefits are most likely. Additional analyses of the real-world data made the identification of prognostic factors possible.

Performance status is the most important prognostic factor. Several multivariate analyses showed ECOG 0–1 to be the most significant predictor of clinical benefit [6,7,13,30,31]. An analysis that focused on negative prognostic factors in response to nivolumab therapy clearly identified the following risk factors of early death: ECOG ≥ 2 (OR 5.66, 95% CI 2.01–15.61; *p* < 0.001), C-reactive protein to albumin ratio >0.3 (OR 10.56, 95% CI 3.61–3086; *p* < 0.001), and poor response to first-line chemotherapy (OR 2.06, 95% CI 1.03–4.14; *p* < 0.001) [31]. Additionally, many authors suggest that liver metastases, brain metastases, and *EGFR* mutation are negative prognostic factors associated with a higher risk of death. Malignant pleural effusion and a high number of metastatic sites were also identified as negative prognostic factors [30,32]. By contrast, a longer PFS on platinum therapy (>6 months) and an objective response to chemotherapy seem to have positive prognostic value in the case of nivolumab treatment. In terms of patient age, the data are inconclusive.

Blood biomarkers can also be considered in treatment decision-making. The use of the lung immune prognostic index (LIPI) based on the baseline derived neutrophil to lymphocyte ratio (dNLR) and lactate dehydrogenase (LDH) was suggested [33,34]. A high LIPI value was indicated as an independent negative prognostic factor (HR 3.67, 95% CI 1.96–6.86; *p* < 0.0001) [33]. Several other

inflammatory-related markers, such as the neutrophil to lymphocyte ratio (NLR), dNLR, LDH, interleukin 8, and indoleamine 2,3 dioxygenase activity were also found to be important [35,36].

It is noteworthy that some of these prognostic factors are also relevant to chemotherapy [37,38]. The negative prognostic value of parameters such as poor performance status, liver or brain metastases, the number of metastatic sites, and an elevated leukocyte count was demonstrated [37,38].

To summarize—the efficacy and safety of nivolumab in the second-line setting of advanced NSCLC have been established in clinical trials and confirmed in real-world practice. Long-term clinical benefit can be obtained in some patients. Good performance status (ECOG 0–1) is crucial, but other clinical variables such as site and number of metastatic lesions, time to failure of first-line chemotherapy, chemotherapy response status, and specific laboratory results should also be considered. There is a further need to collect data on the efficacy of immunotherapy in real-world clinical practice.

**Author Contributions:** M.K.-W. and D.M.K.—conceptualization, M.K.-W.—original draft preparation, D.M.K. and M.K.—review and editing, M.K.—supervision. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.
