2901 - IFN-γ/IL-10 ratio as predictive biomarker for response to anti-PD-1 therapy in metastatic melanoma patients

Background

Anti-PD-1 antibodies represent nowadays a first-choice therapy for metastatic melanoma patients. Despite impressive results in terms of PFS and OS, a proportion of patients does not respond to anti-PD-1 therapy with an overall poor prognosis. Identification of predictive biomarkers is considered an important unmet clinical need to avoid expensive and potentially harmful drugs in patients who will not respond to them. In the last years, many studies have evaluated the role of cytokines on both blood and tissue samples as predictive biomarkers for immunotherapy, with encouraging results.

Methods

Blood samples from 18 patients with metastatic melanoma treated with anti-PD-1 antibodies as first line therapy were collected at baseline. 8 patients were classified as non-responders (best response: PD excluding pseudo-progression with median PFS of 2 months) and 10 patients as responders (best response: PR or CR, with median PFS of 17 months). mRNA expression levels of the main pro- and anti-inflammatory cytokines were evaluated by Real time quantitative PCR in PBMCs obtained from baseline blood samples. Unpaired two-tailed t-test was used for statistical analysis.

Results

IFN-γ mRNA expression levels were higher in responder patients (p < 0.01) whereas IL-10 levels tended to be higher in non-responders (p > 0.05). Combining data for each patient, we noticed a correlation between higher levels of IFN-γ and lower levels of IL-10 for responders and vice versa. Starting from these findings, we observed that the IFN-γ/IL-10 ratio was higher (median: 43,3 vs 5,2) in responders (p < 0.01), with high negative and positive predictive value (NPV:100% and PPV:91% using a threshold of 18). The main lymphocyte subpopulations producing these cytokines were also identified.

Conclusions

Our data suggest an interesting correlation between IFN-γ/IL-10 ratio and response to anti-PD-1 therapy in melanoma patients. This correlation seems to be stronger than using IFN-γ expression levels alone probably because of the influence of anti-inflammatory cytokines. Since this is an exploratory and retrospective analysis of 18 patients, a larger population should be tested to validate our results.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

Dipartimento di Medicina di Precisione, Università degli studi della Campania Luigi Vanvitelli.

Funding

Has not received any funding.

Disclosure

F. Ciardiello: Advisory / Consultancy: Roche; Advisory / Consultancy: Amgen; Advisory / Consultancy: Merck; Advisory / Consultancy: Sanofi; Advisory / Consultancy: Bayer; Advisory / Consultancy: Pfizer; Advisory / Consultancy: Servier; Advisory / Consultancy: BMS; Advisory / Consultancy: Cellgene; Advisory / Consultancy: Lilly; Research grant / Funding (institution): Bayer; Research grant / Funding (institution): Roche; Research grant / Funding (institution): Merck; Research grant / Funding (institution): AstraZeneca; Research grant / Funding (institution): Amgen; Research grant / Funding (institution): Takeda. T. Troiani: Research grant / Funding (institution): Roche; Research grant / Funding (institution): Sanofi; Research grant / Funding (institution): Merck; Research grant / Funding (institution): Amgen; Research grant / Funding (institution): Servier; Research grant / Funding (institution): Novartis; Research grant / Funding (institution): Bayer. All other authors have declared no conflicts of interest.