The introduction of monoclonal antibodies targeting programmed cell death protein-1 (PD-1), programmed death ligand 1 (PD-L1) and cytotoxic T lymphocyte antigen-4 (CTLA-4) has established a new paradigm in the treatment of non-small cell lung cancer (NSCLC).

A significant breakthrough was achieved when regulatory agencies approved nivolumab, pembrolizumab and atezolizumab for patients with NSCLC whose condition had progressed during or following treatment with platinum-doublet chemotherapy. The approval came following a series of clinical trials comparing the three drugs with docetaxel, the standard of care for NSCLC. In their respective clinical trials, nivolumab, pembrolizumab and atezolizumab achieved better overall survival (OS) results than docetaxel1–4.

The three drugs share a similar safety profile and are better tolerated than docetaxel1–4. However, they do present with toxic effects, particularly associated with the immune system, and have been reported to lead to an increased risk of developing immune-related adverse events (irAEs)5.

The purpose of this study was to evaluate the effectiveness and safety of atezolizumab, nivolumab and pembrolizumab in patients with locally advanced or metastatic NSCLC who had received at least one previous treatment.

This was a retrospective observational study that included patients in a third-level hospital treated for locally advanced or metastatic NSCLC.

The inclusion period went from April 2016 to December 2018.

Inclusion criteria were as follows: age of at least 18 years; stage IIIB and IV NSCLC treated with atezolizumab, nivolumab or pembrolizumab; having received previous antineoplastic treatment for their advanced disease. Patients treated as part of a clinical trial or an expanded access program were excluded.

Patient- tumor- and treatment-related variables were analyzed in every case, as reflected in table 1.

The effectiveness of the treatment was analyzed by means of OS and progression-free survival (PFS) measurements.

Safety was evaluated according to the Common Terminology Criteria for Adverse Events (CTCAE) v5.0 of the National Cancer Institute of the United States.

Data were obtained from the patients’ electronic medical records (Orion-Clinic v11.0) and from the pharmacotherapeutic management application for the onco-hematological patients (Farmis-Oncofarm v3.0)

The study sample included eight patients on treatment with atezolizumab [mean age: 55.6 years (range: 53.3-67.4)], 19 patients treated with nivolumab [mean age: 66.3 years (range: 59.4-73.5)] and 16 patients on pembrolizumab [mean age:59.1 years (range: 54.5-65.9)] (Table 1).

Median values for OS and PFS are reflected in figure 1. Median OS was not reached for atezolizumab or pembrolizumab.

Figure 1.

Kaplan-Meier curves for PFS & OS.

(0,33MB).

With regard to atezolizumab, no statistically significant differences were found between median OS and PFS values and any of the analyzed variables.

As far as nivolumab is concerned, median PFS and OS values were statistically higher in patients with 0-1 metastasis than in those with ≥ 2 metastasis. Median PFS was 16.6 months (95%CI 7.2-26)] in the former group and 0.8 months (95%CI 0-6) in the latter [p = 0.004; HR = 5.5 (95%CI 1.5-20)]. The median OS was not reached for patients with 0-1 metastasis but was 0.8 months (95%CI 0-5.9] for those with ≥ 2 metastasis [p = 0.002; HR = 6.4 (95%CI 1.7-24.7)]. Although not statistically significant, a certain trend was observed towards a longer OS in patients with non-squamous cell NSCLC [(p = 0.278; HR = 2.3 (95%CI 0.5-11.4)].

PFS and OS values were statistically improved in ECOG 0-1 patients on pembrolizumab. PFS in these patients was 11.8 months (95%CI 6.4-17.2) as compared with 1 month (95%CI 0.1-1.8) in ECOG ≥ 2 patients [(p = 0.002; HR = 0.097 (95%CI 0.02-0.6)]. As regards OS, the median for ECOG 0-1 was not reached but was 1 month (95%CI 0.1-1.8) for ECOG ≥ 2 patients [p = 0.002; HR = 0.097 (95%CI 0.02-0.6)]. Although no statistically significant OS differences were found in terms of the time elapsed since the previous treatment, a favorable trend was observed when the period was ≥ 6 months [p = 0.208; HR = 2.5 (95%CI 0.6-11.4)].

On the whole, the effectiveness observed in our study for the three drugs under analysis was higher than in the clinical trials published in the literature.

Specifically, the median PFS recorded in our study for patients treated with atezolizumab, nivolumab and pembrolizumab (Figure 1) was higher than median PFS values observed in pivotal clinical trials: 2.8 months (95%CI 2-6.3) for atezolizumab4; 2.33 months (95%CI 2.1-3.3) for nivolumab in non-squamous cell NSCLC1 and 3.48 months (95%CI 2.1-4.9) in squamous-cell NSCLC2; and 3.9 months (95%CI 3.1-4.1) for pembrolizumab3.

These differences can partly be explained by the different methods used to evaluate the tumor's response. Indeed, while clinical trials generally use the RECIST 1.1. criteria (Response Evaluation Criteria In Solid Tumors)6, the most widely used criteria in clinical practice are the immune-related response criteria (IrRC)7, where any increases in lesion size caused by necrosis or hemorrhage are not regarded as indicative of progression of the disease and where stabilization of the disease is considered as PFS, and thus an activity marker. Another factor that may explain these differences has to do with the time at which response is evaluated. While in clinical trials imaging tests were conducted every six weeks, in our study the interval ranged between 9 and 16 weeks.

Median OS was not reached for atezolizumab or pembrolizumab; median OS for nivolumab (Figure 1) was higher than the values obtained in pivotal studies for squamous-cell NSCLC [9.2 months (95%CI 7.3-13.3)] and similar to those for non-squamous cell NSCLC [12,2 months (95%CI 9.7-15)]. The distribution of pulmonary histology in our study did not warrant this difference given that it was precisely patients with squamous-cell NSCLC who obtained lower OS values in clinical trials (68.4% in our study).

No differences were found between this study and pivotal clinical trials in terms of median patient age, percentage of patients ≥ 70 years, ECOG score, and/or percentage of stage IV NSCLC patients that could explain these differences.

Our PFS and OS results were not consistent across all patient subgroups. Values were significantly higher in ECOG 0-1 patients treated with pembrolizumab, which is in line with the findings of other published studies8,9.

As regards nivolumab, both medians were statistically significantly higher for patients with 0-1 metastasis, which coincides with the results published by Garde et al.10.

The frequency and characteristics of adverse events were similar to those in the literature.

Two of the patients treated with nivolumab developed vitiligo. One of them was still being treated three years after the onset of symptoms. In the other, treatment had to be discontinued as a result of grade 3 nephrotoxicity. Four months after discontinuation, the patient developed facial vitiligo, which has progressively worsened. Nearly two years later, the disease remains radiologically stable. The significance of these findings lies in the fact that very few published reports describe the development of vitiligo in NSCLC patients treated with immune therapy11–13, all of them related with a long-term treatment response.

The main limitations of this study are related to the small patient sample and to its single-centric and retrospective nature.

The analyzed sample showed that the effectiveness of nivolumab is lower in patients with ≥ 2 metastases, and that the effectiveness of pembrolizumab is lower in ECOG 2 patients. Development of vitiligo as an adverse event seems to be related to a more durable response to treatment.

No funding.

The authors would like to thank the preventive medicine service of the Alicante General University Hospital for their invaluable help with the statistical analysis of data.

No conflict of interest

The results of the present study were presented in a summarized version at the 13th Meeting of the Spanish Lung Cancer Group, held in Valencia in November 2019.