Abstract

9571 Background: Checkpoint inhibitors have revolutionized the treatment of stage IV melanoma patients. Selection of patients for PD-1 monotherapy or CTLA4/PD1 combination remains an important challenge. We set out to perform a discovery study of pretreatment serum protein biomarkers to identify predictors of progression free survival (PFS) for ipilimumab (IPI) or ipilimumab/nivolumab (IPI/NIVO). Methods: We performed an exploratory analysis of baseline serum samples from 135 treatment-naive patients with metastatic melanoma included in the randomized phase II clinical trial, CheckMate 069 (NCT01927419). We used the RayBiotech 440 human cytokine array and evaluated the relationship of serum protein levels with 44 clinical parameters. R, Prism 7.0 and TensorFlow were used for analyses. Results: We focused on correlation of serum protein markers with PFS as a predictor of long-term benefit. In the IPI arm (n = 46), high FGF4 correlated with worse PFS outcome (p = 0.0012). However, FGF4 levels alone were unable to select responsive vs. non-responsive patients. In contrast, a set of three markers consisting of FGF4 ( < 760pg/ml), CCL15 ( > 2.7 ng/ml), and TACE ( > 600pg/ml) separated non-progressing versus progressing patients. Moreover a small group of FGF4-high patients who were concomitantly TIM-3-low also had longer PFS (combined of both: p = 0.0004, HRlogrank: 0.07, 95% CI: 0.03279 to 0.1533). The same markers did not discriminate between IPI/NIVO patients (p = 0.467, HR: 15). In the IPI/NIVO arm, three different markers could select patients. Patients either with low CCL2 ( < 72 pg/ml) or alternatively with high CCL2 combined with high PDGF-AA ( > 8.2 ng/ml) and low GASP-1 ( < 1.3 ng/ml) had longer PFS (p < 0.0001, HR: 0.115, 95% CI: 0.03848 to 0.3408). Conversely, these markers did not predict benefit for IPI-monotherapy. Conclusions: In this study we identified protein signatures in baseline serum that correlate with PFS for therapies with IPI or IPI/NIVO. The markers were exclusive for IPI or IPI/NIVO but not for both. Additional research is warranted to substantiate these results and evaluate the possibility of incorporating into clinical practice.

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