PD-L1/PD-L2 genomic imbalances in classical Hodgkin lymphoma. Analysis of both Reed Sternberg/Hodgkin and tumor microenvironment cells

MARINA NARBAITZ; IRMA SLAVUTSKY; MARÍA FERNANDA METREBIAN; MAURO GARCÍA MONTENEGRO; ASTRID PAVLOVSKY

Frankfurt

Congreso; 25th Congress of the European Hematology Association; 2020

European Hematology Association

Background: Classical Hodgkin lymphoma (cHL) is characterized by rare tumor Reed-Sternberg/ Hodgkin (RS/H) cells surrounded by a dense immune microenvironment. Genetic alterations at the 9p24.1 locus are recurrent in cHL, resulting in constitutive activation of both PD-L1 and PD-L2, inducing PD-1 signaling and T-cell exhaustion. There is scarce information about genomic imbalances of this locus in tumor microenvironment (TM) cells surrounding the RS/H cells. Aims: In this study, we have characterized genomic imbalances at the 9p24.1 locus by FISH (fluorescence in situ hybridization) on formalin-fixed paraffin embedded (FFPE) diagnostic biopsies of patients with CHL, both in RS/H and TMcells. Results were correlated with PD-L1 protein expression by immunohistochemistry (IHC) and clinical characteristics.Methods: A retrospective analysis of 28 randomly chosen diagnostic biopsies of 377 patients with cHL enrolled in the GATLA LH05 trial were evaluated. For FISH analysis, the SPEC CD274/PDCD1LG2/CEN9 probe (9p24.1) (ZytoVision, Germany) was used. For each patient, 50 RS/H and 100 TM cells were analyzed. The frequency and of 9p24.1 alterations - polysomy (P), copy gain (G), and amplification (A) - were determined. Target:control probe ratio was defined as: A (≥3:1), G (>1:1 but 50% of RS/H cells with G; GP-G subgroup), while the remaining 47% were rich in P (GP-P subgroup). No cases with only P were observed (Figure 1). Twenty seven of 28 cases (96.4%) showed PD-L1 expression. Score +3was observed in 7/28 (25%) of cases, 6 belonging to AGP group and one from GP-G subgroup. No cases with score +3 in the GP-P subgroup were found. Interestingly, the AGP group showed a significant lower mean age at diagnosis (25 years) than the GP group (35 years) (p=0.031). In addition, mediastinal bulky disease was absent in the GP-P subgroup, but present in 37% of the remaining cases. Interim positron emission tomography (PET) scan performed after 3 ABVD was positive (Deauville score 3-4-5) in 55.5% of cases with A compared to 26.3% of patients without A. No differences in other clinical parameters between cases with and without A were observed. The analysis of TM cells showed chromosome 9 monosomy in 100% of evaluated cases from the GP-P subgroup (mean 26.5%; range: 21-38%)compared to 40% of the remaining patients (mean 25.5%; range: 21-30%).Summary/Conclusion: Our results show association between the increased copy number of the 9p24.1 locus with a lower age at diagnosis of CHL, bulky mediastinal disease, augmented PD-L1 protein expression and positive interimPET3. Interestingly, TM cells from the GP-P subgroup showed a high frequency of chromosome 9 monosomy, suggesting an increased chromosome instability associated to the low PD-L1 protein expression found in this subgroup of patients. A larger number of biopsies will be analyzed to confirm this data.