Modeling Growth Of Pediatric T-ALL In Vivo and In Vitro: Clinical Meaning and Activation Of The NFkB Pathway

Pediatric T-cell acute lymphoblastic leukemia (T-ALL) is characterized by the proliferation of T-cell precursors in various sites, such as thymus, bone marrow, blood, lymph nodes or central nervous system. As T-ALL cells alone do not successfully grow in vitro, xenografts of T-ALL cells into NOD/scid/IL-2R null (NSG) mice and long-term co-cultures of T-ALL cells with stromal cells have been developed to study the biology of T-ALL cells (Armstrong et al, Blood, 2009). However, the growth of T-ALL cells in these two systems is highly variable across T-ALL samples. Moreover, the clinical relevance of both assays and, except for NOTCH pathway activation, the molecular pathways involved in successful in vivo and in vitro growths are still elusive.

The aim of this work was to determine the relationships between clinical, biological and molecular characteristics of human T-ALL at diagnosis and the growth of T-ALL in these two systems.

Human T-ALL blood samples were collected at diagnosis from pediatric or young adult patients with T-ALL. 50,000 T-ALL cells were intravenously injected into NSG mice. Mouse bone marrow samples were collected every 3-4 weeks from day 35 to day 210 post-transplant. Leukemic engraftment was monitored using flow cytometry measuring the % of human CD45+CD7+ leukemic cells. Time to leukemic engraftment (TTL) was defined as the time between T-ALL injection and the detection of ≥20% leukemic cells in at least one mouse. In vitro co-culture growth assay consisted in plating 200,000 cells on MS5 or MS5-DL1 (Armstrong, Blood, 2009) and count every 7 days up to 28 days.

A total of 36 samples were tested of which 22 (61%) engrafted into mice. Global median TTL was 82 days (range, 36-121) defining short (TTL<82 days) and long or no engraftment (TTL>82 days) TTL groups. Patient gender, age, mediastinal involvement or abnormal karyotype had no significant impact on TTL. A trend for a shorter TTL was observed for T-ALL samples with a white blood cell count (WBC) > median WBC = 146 G/L (p =0.06). Samples containing more than 20% of TCRαβ or CD8 positive cells exhibited increased incidence of engraftment (p = 0.049 and p=0.04 respectively) whereas CD34, CD1a, CD4 or sCD3 markers were not significantly correlated with TTL. Unlike samples with TLX1, TLX3 overexpression or NOTCH/FBXW7 mutations, samples with SIL-TAL1 deletion exhibited a shorter TTL (p = 0.0004). The 2-year progression free survival of “short TTL” patients was 72% vs 70% for patients with “longer TTL” or no engraftment (p=0.38). T-ALL samples for which growth could be achieved on MS5 cells also displayed a shorter TTL.

To unravel molecular mechanisms involved in the growth of leukemic cells in these two systems, micro-arrays were performed for 8 “short TTL” T-ALL versus 8 “long TTL or no engraftment” T-ALL. 346 genes were differentially express in short TTL samples compared to long/no TTL samples (P<0.05, fold change: 1.5). As expected, most of genes up-regulated in short TTL group were implicated in cell cycle function enhancing the commitment of cells to S/M phases. Analysis of regulated networks revealed that several indirect modulators of NFkB (MAL, AhR and CYLD) were significantly up/down regulated in short TTL patient samples resulting in NFkB activation.

Overall, T-ALL with SIL-TAL1 deletion display an increased ability to engraft into NSG mice, in accordance with increased WBC in T-ALL patients. Contrary to B-ALL, shorter TTL is not associated with poor prognosis in T-ALL. Moreover, NSG engraftment and co-culture on stromal cells are well correlated. A shorter TTL seems to be associated with an increased leukemic proliferation through NFkB activation.