Blood serum amyloid A as potential biomarker of pembrolizumab efficacy for patients affected by advanced non-small cell lung cancer (NSCLC) overexpressing PD-L1: Early results of the FoRECATT Study.

Abstract

9598 Background: Identifying the patients who may benefit the most from immune checkpoints inhibitors remains a great challenge for clinicians. The tumor-derived Serum Amyloid A (SAA) inhibits the immune-response in melanoma patients. Here we present the early results of FoRECATT study investigating on blood SAA as biomarker of response to upfront pembrolizumab in patients with advanced non-small-cell lung cancer (NSCLC). Methods: In this prospective study, patients with PD-L1 ≥ 50% receiving upfront pembrolizumab (P cohort) and with PD-L1 0-49% treated with chemotherapy (CT cohort), were evaluated for blood SAA and radiological response at baseline and every 9 weeks. Primary endpoint was response rate (RR) according to Response Evaluation Criteria in Solid Tumors 1.1; secondary endpoints were progression-free (PFS) and overall survival (OS). The most accurate SAA cut-off to predict response was established with ROC-analysis in the P cohort. Results: In the P Cohort (n = 42), the overall RR was 38%. After a median follow-up of 18.5 months (mo), baseline SAA ≤ the ROC-derived cut-off (29.9 mg/L; n = 14/42, 33%) was significantly associated with higher RR (53.6 versus 7.1%; OR 15, 95%CI 1.72-130.7, P= 0.009), longer PFS (17.4 versus 2.1 mo; P< 0.0001) and OS (not reached versus 7.2 mo; P< 0.0001) compared with SAA > 29.9 mg/L. In multivariate analysis, low SAA positively affects PFS ( P= 0.001) and OS ( P= 0.048) irrespective of ECOG PS, number of metastatic sites and pleural effusion. SAA monitoring (n = 40) was also significantly associated with survival endpoints: median PFS 17.4 versus 2.1 mo and median OS not reached versus 7.2 mo when SAA remained low (n = 14) and high (n = 12), respectively. In the CT Cohort (n = 30), RR was not significantly affected by SAA level, while low SAA at baseline (n = 17) was associated with better PFS (HR = 0.42, 95%CI 0.16-1.10, P= 0.02) and OS (HR = 0.16, 95% CI 0.04-0.55, P= 0.0004). Conclusions: Low SAA predicts a higher likelihood of response to upfront pembrolizumab only and good survival outcomes irrespective of treatment in advanced NSCLC patients. Therefore, a simple blood test might be useful to identify patients likely to derive better outcomes from immunotherapy. A further study (FoRECATT-2) is ongoing to confirm the results in a larger validation cohort and to assess the potential effect of SAA on immune response in vitro assays.