Comprehensive Analysis of the Tumoral and Microenvironmental Transcriptome of Lymph Nodes and Bone Marrow in Mantle Cell Lymphoma

The Reverse Transcription Multiplex Ligation-dependent Probe Amplification (RT-MLPA) technique is a method allowing for the semi-quantitative concomitant analysis of a number of transcripts of interest in cells or tissues. It relies on the amplification of complementary DNA (cDNA) derived from messenger ribonucleic acid (mRNA) after binding and ligation of specific probes on adjacent regions. It is thus applicable to the relatively fragmented low-quality RNA that can be derived from formalin-fixed paraffin embedded samples (FFPE), ill adapted to other methods such as RNASeq. Adequate tagging provides specific amplicon sizes for each marker, thus becoming readily identifiable after capillary electrophoresis. Signal normalization further allows to compare relative levels of transcripts in a given run.

Here we report on the application of this method to explore the tumoral and microenvironmental, i.e. intrinsic and extrinsic, transcriptome of mantle cell lymphoma (MCL). Data were obtained from i)bone marrow (BM) and lymph nodes (LN) from normal individuals, ii)BM and LN from MCL patients at diagnosis (n=33) and iii)BM and LN from relapsing MCL patients (n=19). Samples were either frozen BM and frozen of FFPE LN. mRNAs were extracted and reverse-transcripted in cDNA before being processed for MLPA investigating for 17 transcripts of interest. The latter explored the intrinsic tumoral expression of CCND1, SOX11 and MKI67. The extrinsic parameters of the microenvironment investigated concerned the monocyte/macrophage compartment (CD14, CD163), T-cells (CD3E/CD3, CD8A/CD8), NK cells (CD94), the costimulation molecule CD40-ligand (CD40LG/CD40L) and immune-checkpoint molecules (CD152/CTLA4, PDC1/PD1, CD274/PDL1, PDCD1LG2/PDL2, INDO1/IDO) as well as cytokines (CSF1/MCSF , IL10, TGFB1).

The samples studied were 39 MCL LN, 20 MCL BM, 10 non-tumoral LN and 9 normal BM.

RT-MLPA readily differentiated MCL samples from non-tumoral samples by showing high levels of CCND1 and SOX11 transcripts in patients. Extrinsic markers expression, analyzed in principal component analysis (PCA) clearly segregated diseased and normal LN, while the difference was less clear in BM. Macrophage markers transcript levels were lower in patients' LN which also showed a down regulation of global T- and NK-cells yet increased levels of CD8 transcripts. PDL1 expression was higher in MCL BM compared to controls.

LN from MCL patients with aggressive characteristics displayed lower CCND1 but higher MKI67 transcripts than other MCL patients' samples. These aggressive forms were found to be significantly associated with high CD14 transcripts in LN and high CD163 in BM. In addition, blastoid morphology was significantly associated with high CD14 (p=.005) and CD163 transcript levels in LN. Regarding T-cells, MCL aggressive forms had significantly lower amounts of CD3E transcripts, yet increased CD8 expression.

LN from patients with a high proliferative index displayed higher levels of CD8 but also of CTLA-4, PD-1 and PDL1, suggesting that these T-cells are exhausted.

A significant upregulation of the levels of CCND1 transcripts was observed in relapse LN. Conversely, CD163, CD3E, CD40L and CD94 levels were lower in relapsed samples compared to diagnosis, which could reflect immune depletion secondary to immuno-chemotherapy.

This study confirms, in an original transcriptomic approach, the great heterogeneity of the MCL microenvironment. Besides being a useful tool at diagnosis, RT-MLPA successfully allowed for a concomitant approach of a large number of transcripts and thus a precise evaluation of intrinsic prognostic and extrinsic features of MCL. Important, yet seldom described, modulations of T-cells, NK cells, macrophages and cytokines were observed, with different profiles segregating MCL with aggressive morphology and those with resistance to treatment.