PS1304 FOLLICULAR HELPER T‐CELLS FORM MUTUALLY SUPPORTIVE INTERACTIONS WITH FOLLICULAR LYMPHOMA B‐CELLS THAT MAY SUPPORT TUMOUR GROWTH

Background: Follicular lymphoma (FL) is a germinal centre (GC) B‐cell neoplasm that is dependent on interactions with its non‐malignant tumour microenvironment (TME). Follicular helper T‐cells (Tfh) play a pivotal role in supporting healthy GC B‐cell development, proliferation, and maturation. In FL, Tfh are present in similar proportions to reactive lymphoid tissue (Townsend, ASH 2014 abs 144) and can support survival of FL cells in vitro (Ame Thomas, Blood 2015). It remains unclear to what extent Tfh are able to drive FL growth and transformation. Aims: To explore whether mutual FL‐T FH interactions influence the survival and phenotype of both FL cells and infiltrating T FH . Methods: Disaggregated FL lymph node (LN) tissue and fine needle aspirates were obtained from untreated or relapsed FL patients (n = 19). Total FL LN cells (n = 10) were cultured unmodified, or with depletion of either T FH or B‐cells with immunomagnetic anti‐CD3 or anti‐CD19 beads (n = 10), respectively. To study in situ spatial interactions, 4‐colour confocal immunofluorescence microscopy was performed on archival FL tissue (n = 20). Results: In FL LN, T FH (PD‐1 hi ICOS + CXCR5 + ) comprise 28.6% (95% CI: 21.4–35.7) of CD4+ T‐cells by flow cytometry. Only a minority (5.0%; 95% CI: 0.6–9.4) had a FoxP3 + regulatory phenotype. After CD19 (FL cell) depletion, T FH partially lost their phenotype, with reduced CXCR5 and PD‐1 expression, as well as increased T FH cell death. There was no change in T FH ICOS expression after CD19 depletion, whilst ICOS‐L was upregulated in FL cells after CD3 (T FH ) depletion. This partly mirrors findings with healthy T FH and B‐cells), demonstrating mutually beneficial and dynamic T FH ‐FL interactions. These findings were confirmed in cultures with flow‐sorted T FH and FL B‐cells (n = 6). FL cells had higher expression of B‐cell activation markers (CD86 and HLA‐DR) when cultured in the presence of flow‐sorted T FH compared to those without. Additionally, we observed a reciprocal increase in T FH activation, with higher CD69 expression (but not CD25 or Tim3) in those cultured with FL cells. These findings were specific to B‐T FH interactions and were either not seen or diminished when FL cells were cultured with non‐T FH CD4 T‐cells. In FL tissue, T FH within closely interacted and correlated in number with proliferating FL cells: 58.5% Ki67 + (95% CI: 50.8–66.2) cells were in contact with T FH . This mirrored findings in reactive lymph node tissue (n = 5) and suggests the number of T FH cells may determine FL proliferation rate. We also found a close spatial relationship between T FH and FL‐cell expression of AID and MYC, both of which have been implicated in FL progression and transformation. Further characterisation of these complex cell populations within the FL TME requires highly multiplexed imaging techniques. We have designed an imaging mass cytometry panel to enable simultaneous assessment of 23 antigens in the same FL tissue section. We show that IMC is a feasible method for visualising T FH in FFPE tissue, allowing multiparameter analysis with surrounding FL cells and other TME cells. Summary/Conclusion: Here we provide evidence that T FH form mutually supportive, reciprocal interactions with FL B‐cells, implying a key role in promoting FL growth. T FH have a close relationship with FL proliferation and markers suggesting genomic dysregulation, which may drive FL progression. In an immunotherapy era, better understanding of these interactions will suggest future targets and pathways for therapeutic intervention in FL.