Baseline and early changes in circulating Serum Amyloid A (SAA) predict survival outcomes in advanced non-small cell lung cancer patients treated with Anti-PD-1/PD-L1 monotherapy.

ABSTRACT

BACKGROUND: Systemic inflammation plays an important role in carcinogenesis and is associated with overall survival in patients with different cancer types, including those treated with immune checkpoint blockade (ICB). Serum Amyloid A (SAA) is an acute-phase protein and a marker of persistent inflammation. We hypothesized that circulating SAA may predict outcomes in advanced non-small cell lung (aNSCLC) patients treated with PD-1/PD-L1 ICB. MATERIALS AND METHODS: This retrospective study included 91 aNSCLC patients who received anti-PD-(L)1 monotherapy in Sun Yat-sen University Cancer Center (Guangzhou, China) between August 2016 and June 2018. We examined the impact of circulating SAA at baseline and 8 (±2) weeks later on overall survival (OS). X-tile program was used to determine the cut-off values which optimized the significance of the split between Kaplan-Meier survival curves. Kaplan-Meier methodology and Cox regression analyses were conducted for survival analyses. RESULTS: The optimal cut-off value of baseline SAA for OS stratification was 137.6 mg/L. In univariate analysis, both high level of baseline SAA (hazard ratio [HR], 2.76; 95% confidence interval [CI], 1.47-5.18; P = 0.002) and lack of early SAA descent (HR, 1.51; 95% CI, 1.11-2.06; P = 0.009) were significantly associated with inferior OS. In multivariate analysis, gender, smoking status, performance status, liver metastasis, neutrophil-to-lymphocyte ratio, baseline SAA and early changes in SAA independently predicted OS (all with P < 0.05). A combined baseline SAA ≥ 137.6 mg/L and without early SAA descent identified a small cohort with remarkably worse OS (median, 3.2 months). CONCLUSIONS: Both high baseline and lack of early decline in circulating SAA are significantly associated with inferior outcomes in aNSCLC patients treated with PD-1/PD-L1 ICB. Combined these two SAA indexes provided improved risk stratification. The prognostic value of this simple, readily-available, and cost-effective biomarker warrants larger, prospective validation before definitive recommendation can be made.