Abstract

e20543 Background: Nivolumab (NIV) is now widely used in ≥ 2ndline of advanced non-small cell lung cancer (NSCLC). The aims of this real-life study were (1) to evaluate its efficacy, (2) to identify predictive factors of worsening under NIV. Methods: This retrospective multicentric study (8 Thoracic Oncology regional centers) included all patients treated by NIV in 2015 (follow-up until 2016, June 30th). Demographics and tumor characteristics were extracted. Objective tumor response was assessed by computerized tomography (RECIST v1.1). Median NIV treatment duration and overall survival (OS) were calculated. Clinical and biological factors evaluated as predictive markers of earlyprogression (discontinuation before 1stevaluation) were: NSCLC histology, PS, smoking-history, number of previous chemotherapy received and initial albumine (alb), C-reactive protein (CRP) and lactate dehydrogenase (LDH) serum levels. For these factors, patients quickly worsened and responders were compared (Student’s t-test, Chi-2 test). Results: Our cohort included 297 patients (non-squamous NSCLC n = 181 [61%]; PS > 1 n = 66 [22%]). At first evaluation, partial response was observed in 16% (n = 47), stable disease in 18% (n = 54) and progression in 49% (n = 144). A total of 45 patients (15%) were early progressors and no evaluation could be performed. Median NIV duration of treatment was 2.8 months [0-15], OS 6.9 months [0-16] (n = 146 still alive on June 30th, 2016). Early progressor patients were PS > 1 in 42% versus (vs) 21% for responders (mean PS 1.7 vs 1.0 for responders (p < 0.001)) and an alb level of 29.9 g/L vs 33.6 (p = 0.03). Mean CRP (p = 0.06) and LDH (p = 0.08) serum levels, NSCLC histology (p = 0.64), smoking-history (p = 0.08), number of previous chemotherapy (p = 0.90) were non-significant. Conclusions: This study confirms that NIV may be effective in NSCLC, with a 16% objective response rate. However early progression observed in 15% of patients was associated with a lower initial alb level and a higher PS. Missing data for LDH and CRP levels may have influenced their non-significance. Further studies must be realized to determine the interest of these factors as predictive markers.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call