PD-L1 expression with immune-infiltrate evaluation and outcome prediction in melanoma patients treated with ipilimumab

G. Madonna, C. Ballesteros-Merino, Z. Feng, C. Bifulco, M. Capone, D. Giannarelli, D. Mallardo, E. Simeone, A. M. Grimaldi, C. Caraco, G. Botti, B. A. Fox, P. A. Ascierto

Research output: Contribution to journalArticlepeer-review


Background: Tumor microenvironment may have a key role in providing immunological markers that can help predict clinical response to treatment with checkpoint inhibitors. We investigated whether the baseline expression of PD-L1 in advanced melanoma patients treated with ipilimumab may correlate with clinical outcome. Methods: PD-L1 expression was assessed in 114 patients with advanced melanoma treated with ipilimumab and, in a cohort of 77 patients, a comprehensive assessment using multispectral imaging to assess the presence and distribution of CD3+, CD8+, CD163+, FOXP3+ and PD-L1+ cells inside and at periphery of the tumor was performed. Results: PD-L1 status alone was not a predictive biomarker for response or survival. There was an association between clinical benefit from ipilimumab therapy with the coexistence of low densities of CD8+ and high densities of CD163+ PD-L1+ cells at the periphery of the tumor. Conclusions: To explain the association of this peculiar microenvironment with clinical benefit from ipilimumab, we proposed a model where baseline CD8 cells levels are low due to inhibitory effect of Tregs and to pro-tumor activity of TAM M2 (CD163+ PD-L1+ cells). Ipilimumab treatment causes a decrease of Treg cells, mediated by ADCC from macrophages, with a concomitant change in TAM polarization that switches from M2 to M1 with a subsequent attraction of CD8 cells and the increase of antitumor response.
Original languageEnglish
Pages (from-to)e1405206
Issue number12
Publication statusPublished - Sep 5 2018


  • Immunotherapy
  • Ipilimumab
  • Melanoma
  • PD-L1
  • Tumor microenvironment


Dive into the research topics of 'PD-L1 expression with immune-infiltrate evaluation and outcome prediction in melanoma patients treated with ipilimumab'. Together they form a unique fingerprint.

Cite this