Abstract Non-Hodgkin lymphomas are a heterogeneous group of B-cell neoplasms that can present with nodal and extranodal (EN) disease. The presence of the MYD88L265P mutation is associated with EN disease and inferior outcomes in various B-cell lymphomas, including lymphomas of immune-privileged sites such as the central nervous system (CNS). However, the mechanistic link between MYD88L265Pand EN localization remains poorly understood. This study aims to determine whether MYD88L265P directly contributes to EN disease by (i) providing a growth advantage in stimulus-poor environments and/or (ii) favoring EN tropism.To address these questions, we established a human B-cell lymphoma model system for functional ex vivo and in vivo studies. Primary germinal center B-cells were immortalized by overexpression of BCL2 and MYC, enabling long-term culture on follicular dendritic cells (YK6) expressing CD40L and IL-21, modeling a nodal microenvironment. We used CRISPR/Cas9 to knock out CDKN2A, as CDKN2A is frequently lost in EN lymphomas, particularly CNS lymphomas. Finally, we retrovirally expressed MYD88wt or MYD88L265P. Cells were stably transduced with enhanced firefly luciferase to allow bioluminescence imaging in vivo.(i) To study whether MYD88L265P supports lymphoma cell growth in EN environments, we assessed ex vivo cell growth after YK6 removal. Unlike MYD88wt cells, MYD88L265P cells showed sustained growth without YK6. Next, we mixed MYD88wt (blue fluorescent protein (BFP)-positive) and MYD88L265P (BFP-negative) cells at equal ratios and performed competitive growth assays. Interestingly, the MYD88wt cells had a selective advantage on YK6 feeder cells, whereas MYD88L265P outgrew MYD88wt cells after YK6 removal. To test whether this MYD88L265P-mediated independence from external stimuli also confers a growth advantage in vivo, NOD-scid IL2Rgammanull (NSG) mice were injected subcutaneously with MYD88wt cells into the right flank and MYD88L265P cells into the left flank (N=9, i.e., 3 human donors, 3 replicates each). While MYD88wt cells caused tumors in 4 out of 9 mice, MYD88L265P cells formed significantly larger tumors in all 9 mice, and required euthanasia in every case (0.80 ± 0.14 cm vs. 1.24 ± 0.24 cm, P=0.006). Currently, we are conducting intracranial injections of MYD88wt or MYD88L265P cells in NMRI-Foxn1nu/nu mice to assess whether this growth advantage is preserved in the CNS.(ii) To study if MYD88L265Palso directly promotes EN tropism, we performed a competitive transplant experiment with intravenous injections of a 1-to-1 mix of MYD88wt (BFP-positive) and MYD88L265P (BFP-negative) cells into NSG mice (N=24, i.e., 3 human donors, 4 male and 4 female mice each). This immunocompromised mouse model allows the investigation of EN tropism independent of potential immune escape mechanisms. The experiment has been completed for 14 mice, whose lymphoid and non-lymphoid organs were analyzed for the presence of human lymphoma cells. Most commonly infiltrated organs were the bone marrow (12/13, 92%), ovaries (3/6, 50%), brain (6/14, 43%), eyes (3/8, 38%), and testes (1/6, 17%). EN tumors were found in 10 out of 14 mice (71%), 6 of which were subcutaneous (43%). Infiltration of the kidneys was detected in 1 mouse (1/13, 8%). Interestingly, the spleen was not infiltrated in any of the mice (0/13, 0%). In all tumors and infiltrated organs, 99% of lymphoma cells carried the MYD88L265P mutation, confirming our previous finding of a selective advantage in stimulus-poor environments. The experiment is ongoing for 10 animals. As this model recapitulates many of the EN manifestations of human MYD88L265P mutant lymphomas, we are currently conducting whole transcriptome sequencing to understand the underlying biology of this selective tropism to distinct EN sites.In summary, we show that MYD88L265P renders B-cell lymphoma cells independent of external stimuli, leading to an EN growth advantage. Moreover, our model recapitulates many patterns of EN involvement in human MYD88L265P mutant lymphomas, suggesting that the L265P mutation is directly involved in mediating this tropism to specific EN sites. Deciphering the mechanisms by which MYD88 mutations contribute to EN disease will help to develop targeted approaches to more effectively treat and potentially prevent EN disease.
Vick et al. (Mon,) studied this question.