Background: The advent of immune checkpoint inhibitor (ICI) has revolutionized the therapeutic arsenal for metastatic non-small cell lung cancer. However, better therapeutic options after a progression on frontline ICIs are urgently needed and remain to be defined. ICI retreatment (rt) may constitute a rational therapeutic alternative in these patients. However prospective studies evaluating such opportunity are still lacking. Methods: We retrospectively collected data from 30 metastatic NSCLC patients (pts) having progressed on ICI (anti-PD-1, anti-PD-L1 or anti-PD-1/anti-CTLA4 combined or not with histologydriven chemotherapy) who were retreated by double-agent ICI (ipilimumab and nivolumab) combined or not with histology-driven ChT in order to evaluate rt overall survival (OS) and rt progression-free-survival (PFS). The rt PFS was calculated from rt ICI initiation to death, last follow-up or disease progression. An assessment of correlation between certain clinical variables like sex, age, smoking history and others; and survival variables was performed. Results: The median age of the 30 pts included in the analysis was 64 years (63% of pts were < 70 years old and 37% ≥ 70 years old) and 53% (n = 16) of the pts were male. Most suffered from a non-sq NSCLC (73%), with bone mets (66%), and were current or former smokers (90%). Median OS and median PFS in the whole group were 9 months (m) [95% confidence interval (CI): 5.7 - 14] and 2.5 m [95% CI: 2 - 4.6] respectively. Retreatment PFS was higher in pts having received dual ICI combined with ChT than dual ICI only [95% CI: 4.3 (3.7 – 6) versus 2.1 (1.5 – 2.9) m; hazard ratio (HR): 0.51 (0.24 – 1.10); p – value (p) = 0.085]. Retreatment OS and PFS were numerically higher in pts having received single- or double- ICI as prior therapy compared to ICI plus ChT [9.6 (6.4 – not reached) m versus 8.4 (3.7 – 13.5) m; HR: 0.53 (0.22 – 1.29), p = 0.15], and [4.2 (2.2 – not reached) m versus 2.4 (2 – 4.5 m); HR: 0.56 (0.25 – 1.24); p = 0.14), respectively], regardless of the rt. When we assessed the correlation between prior therapy and rt, we found the association of ChT to dual-ICI after prior single- or dual-ICI as first line treatment leads to a significantly higher PFS [4.5 (4 – NR) m versus 2.5 (1.6 – NR) m] compared to pts having received first ChT plus ICI then dual-ICI alone as second line treatment [HR 0.35 (95% CI: 0.13 – 0.93); p: 0.03]. The trend in OS benefit probably also favored ChT with dual-ICI as rt after prior single- or dual-ICI (HR 0.55; p = 0.23). Binary variables such as sex, histology, PD-L1 expression, smoking history, therapeutic discontinuation interval (TDI), baseline derived neutrophil-to-lymphocyte ratio dNLR) and baseline absolute lymphocyte counts (ALC) influenced positively the PFS in pts receiving dual ICI combined with ChT [ sex HR: 0.28; non-squamous histology HR: 0.26; PDL1 expression HR:0.22; former smoker HR: 0.23; never-smoker HR: 0.40; TDI less than 30 days HR: 0.09; high (≥ 3) dNLR HR: 0.22; and low (<1000) ALC HR: 0.37]. Conclusions: Patients previously treated by ICI could be retreated by ICI combined with histology-driven chemotherapy in an individual approach, under certain circumstances, which are yet to be formally evaluated and determined.