P78.06 Dynamic Risk Prediction for Disease Control With Nivolumab in Advanced or Recurrent Non-Small Cell Lung Cancer Patients (NewEpoch)

Abstract

Immune checkpoint inhibitors (ICI) have been widely used for treatment of advanced or recurrent non-small cell lung cancer (NSCLC), but prediction of their efficacy remains difficult before and at early phases of therapy. Here, we aimed to clarify early clinical predictors for disease control with nivolumab in patients with NSCLC. We prospectively collected a cohort of patients with advanced or recurrent NSCLC who received nivolumab every 2 weeks as second or third-line treatment from Aug 2016 to Dec 2017. Disease control was defined as continuing CR/PR/SD according to the RECIST at 25 weeks after the start of nivolumab. QOL score by EQ-5D-5L was collected at baseline and at weeks 5, 9, 13 and 25 (after 12 cycles). Potential clinical biomarkers included patient characteristics, laboratory data, performance status (PS) and QOL score before and at nivolumab week 9 (after 4 cycles), and immune-related adverse event (irAE) at week 9. 243 patient cohort comprised of 50 (21%) females, with mean age of 67.7 years old, 91% good PS. 33% had squamous cell carcinoma, 67% had non-squamous NSCLC. 66% had PD-L1 1% or more expression and 88% were current or past smokers. At this data cutoff, the disease control rate (DCR) and the objective response rate (ORR) at week 25 were 41.2% (95% CI, 34.9%–47.6%) and 18.5% (95% CI, 13.8–24.0) respectively. Median progression-free survival (PFS) was 3.9 months (95% CI, 3.3–5.5) and overall survival (OS) was 13.0 months (95% CI, 11.4–16.5). 9% (22/243) experienced grade 3 or more irAE at week 9, including 5% (12/243) pulmonary toxicity. The multivariate analyses identified male gender [odds ratio (OR), 0.29; 95% CI, 0.13–0.65, p=0.0027] and disease control at week 9 (OR, 11.8; 95% CI, 3.4–40.7; p<0.0001) as predictors for disease control at week 25. Similarly, high lymphocyte count at baseline (p=0.03) and at week 9 (p=0.016) and low serum creatinine value at week 9 (p=0.028) were associated with overall response at week 25. Grade 3 or more irAE at week 9 was not associated with disease control (OR, 1.45; 95% CI, 0.85–2.45; p=0.17) and overall response (OR, 1.16; 95% CI, 0.59–2.27; p=0.66) at week 25. Landmark analysis of overall survival starting from week 25 (n=152) showed that status of disease control, PS (0, 1 vs. 2 or more, p=0.0051) and QOL score such as pain (p=0.0086) or anxiety (p=0.0039) at week 25 were independent prognostic factors as well as gender (p=0.012), nivolumab as the third line therapy (p=0.0042). Body mass index (p=0.017) and co-morbidity such as interstitial lung disease, chronic obstructive pulmonary disease and liver disease at baseline (p=0.0028) were independently associated with pulmonary toxicity by nivolumab. Gender, lymphocyte count and serum creatinine at week 9 may be clinical predictors for nivolumab efficacy in patients with advanced or recurrent NSCLC. In addition, patient reported outcomes may be independent prognostic factors.