Poster Presentation 27th Lorne Cancer Conference 2015

Associations between mechanisms of BRAF inhibitor resistance and PDL1 expression in BRAF inhibitor treated patient progressed lesions  (#309)

James S Wilmott 1 2 , Hojabr Kakavand 1 2 , Alex M Menzies 1 2 , Ricardo Vilain 1 3 , Helen Rizos 4 , Gulietta M Pupo 5 , Matteo S Carlino 1 5 , Georgina V Long 1 2 , Ricahrd A Scolyer 1 3 6
  1. Melanoma Institute Australia, Camperdown/Sydney, NSW, Australia
  2. Medicine, The University of Sydney, Sydney, NSW, Australia
  3. Tissue Pathology and Diagnostic Oncology , Royal Prince Alfred Hospital, Sydney, NSW, Australia
  4. Sydney Medical School, Macquaire University, ,Sydney, New South Wales, Australia
  5. Centre for Cancer Research, The University of Sydney at Westmead Millennium Institute, Sydney, NSW, Australia
  6. Pathology, The University of Sydney, Sydney, NSW, Australia

Purpose: To evaluate the association between the immunoreactivity of the immune inhibitory ligand PD-L1 with tumour infiltrating lymphocytes (TILs) and known mechanisms of MAPK inhibitor (MAPKi) resistance in tumor biopsies resected at the time of MAPKi disease progression.

Experimental Design: 23 patients with disease progression on MAPKi (vemurafenib, dabrafenib, and combination dabrafenib and trametinib) each had a progressing tumor biospied. All tumors underwent MAPKi resistance screening including RNA expression profiling. Morphological features, PD-L1 immunohistochemical immunoreactivity, and RNA expression profile were correlated with mechanisms of resistance (including mutations in NRAS and MEK, and amplification and aberrant splicing of BRAF).

Results: PD-L1 immunoreactivity was significantly higher in tumors that harbored an aberrant BRAF splicing variant (n=8) or MEK mutations (n=5) than those with BRAF amplification (n=4), NRAS mutations (n=2) or those with unknown resistance mechanisms (n=4) (mean 27% vs 1%, p=0.049). Additionally, the PD-L1 immunoreactivity in tumors with known MAPK-reactivating mechanisms (n=19) tended to be higher than those with unknown resistance mechanisms (n=4) (mean 17% vs 0%, p=0.283). The immunoreactivity of PDL1 in progressed lesions was independent of BRAFV600 genotype, RECIST response and type of MAPK inhibitor. Correlations with TIL subsets, RNA expression, and response to subsequent anti-PD-1 therapy are underway.

Conclusions: The induction of PD-L1 immunoreactivity following MAPKi treatment in human melanoma is variable, however specific mechanisms of MAPKi resistance may confer increased expression of the inhibitory ligand. Further studies are underway to examining the association of the mechanisms of resistance and PDL1 expression with TIL infiltration, the RNA expression profile, and the response to subsequent anti-PD1 therapy.