Abstract 1039P
Background
Immunotherapy with anti-PD1/PDL1 antibodies has become a standard treatment for advanced NSCLC, with higher efficacy in patients with PDL1 positive tumors. However, the difficulty to achieve tumor samples for PDL1 testing requires the search of new biomarkers. This study aims to assess the utility of circulating biomarkers for predicting response to anti-PD1/PDL1 therapies in NSCLC.
Methods
Blood samples from 50 NSCLC patients (pts) were collected before second-line treatment with anti-PD1/PDL1 antibodies. Plasma biomarkers’ levels were measured by Multiplex bead-based assays (ProcartaPlex, Thermo Fisher). Non-parametric tests were used to correlate clinico-pathological characteristics with analytical variables dichotomized using median as cut-off. To evaluate survival, Cox Regression analysis and Kaplan Meier curves with log-rank test were performed.
Results
Soluble sCD137, sGITR and sIDO correlated significantly with response (p=0,016, p= 0,018, p=0,030, respectively), while sPDL1 showed a trend. Three groups of pts were established according to high or low levels of these 4 markers as following: G1, pts with low levels of 3 or 4 markers; G2, pts with low levels of 2 and high levels of 2 markers; G3, pts with high levels of 3 or 4 markers. G1 pts showed significant lower response rate (p= 0.011), and G3 patients had better OS (NR vs. 8.6 vs. 6.33 months, p= 0.010). Pts with higher sPDL1 had a significant increase in PFS (NR vs. 3 months, p=0.017), while a trend toward an increase in OS (NR vs. 8.27 months, p= 0.081) compared to those with lower levels. Multivariate analysis for OS revealed that combination of 4 soluble markers has an independent prognostic value in this cohort (HR=0.595, p= 0.04).
Conclusions
Circulating biomarkers can be reliable detected in plasma of advanced NSCLC pts. sCD137, sIDO, sGITR and sPDL1 seem to be related to anti-PD1/PDL1 response or PFS, and combination of these soluble markers may be an independent prognostic factor for OS. Supported by grant PI18/00226 from ISCIII, CB16/12/00350 from CIBEROnc, F. Arnal Planelles and AMACMA.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
FIHGUV-Fundación de Investigación del Hospital General Universitario de Valencia.
Funding
PI18/00226 from ISCIII, CB16/12/00350 from CIBERONC, F. Arnal Planelles and AMACMA.
Disclosure
All authors have declared no conflicts of interest.