Longterm Serial Transplantability of BCR-ABL+ Chronic Phase CML Cells Is Predicted by a 6-Week LTC-IC Assay.

Abstract Abstract 2770 Since the advent of tyrosine kinase inhibitors (TKIs) to treat CML patients, the development of methods to measure and characterize the CML stem cell compartment has stimulated increasing interest. Here we compared different methods for quantifying the relative frequency of very primitive Ph+/BCR-ABL+ in 33 chronic phase CML patients that had not been previously treated with TKIs. Materials and Methods Longterm culture-initiating cells (LTC-ICs) were assayed by coculturing immunomagnetically isolated CD34+ cells for 5-weeks in LTCs containing murine stromal cells (n=10) and/or for 6-weeks in LTCs containing murine stromal cells secreting human SCF, IL-3 and G-CSF (n=33). Genotyping of the LTC-ICs was based on genotyping individual colonies generated in methylcellulose assays of cells harvested at the end of the 5 or 6 weeks of coculture, either by karyotyping G-banded metaphases (n=16) or by qRT-PCR of extracted RNA (n=17). In vivo assays were performed on 2 samples by injecting 106 CD34+ cells (>97% and 44% Ph+ 6-week LTC-ICs) IV into primary sublethally irradiated NOD/SCID IL2Rγc−/− (NSG) mice and after 30–35 weeks further into secondary NSG mice. All mice were analyzed periodically for human hematopoietic cells by flow cytometry of marrow aspirates. CD34+CD38− cells isolated by FACS from primary CML samples (n=17) were spotted on slides and examined by FISH for the presence of the BCR-ABL gene. Results The proportion of 6-week LTC-ICs that was Ph+/BCR-ABL+ ranged from <5% to 100%. Although the CML LTC-ICs represented >80% of the LTC-ICs in 36% of the 33 cases studied, these represented <50% in 45% of these cases. Ph+/BCR-ABL+ LTC-ICs were more prevalent when measured with the 5-week LTC-IC assay (86±7% Ph+/BCR-ABL+) than with the 6-week LTC-IC assay (11±8% Ph+/BCR-ABL+, n=10). FISH analysis of the initial CD34+CD38- cells showed that >80% of these were BCR-ABL+ in 70% of the 17 cases studied. Notably, these latter values were not correlated with the proportion of leukemic 6-week LTC-ICs in the same samples (Spearman rank correlation r = 0.43, p = 0.08). Primary NSG mice transplanted with CD34+ cells from the patient with no detectable normal LTC-ICs regenerated almost exclusively differentiated human myeloid cells for up to 35 weeks and at increasing levels at the later time points (35% of total marrow cells after 35 weeks). Cells obtained 8 and 35 weeks post-transplant showed these contained readily detectable clonogenic cells which were exclusively BCR-ABL+ (84 genotyped colonies). Secondary recipients were again repopulated with exclusively BCR-ABL+ myeloid cells. Recipients of the second sample showed a transient early peak of myeloid cells followed by a peak of B lymphoid cells at 8 weeks, at which time only 20% of the human CFCs present were BCR-ABL+. At 35 weeks post-transplant, human cells were still detectable in these mice (5±3% of the marrow) and myeloid cells had again become the predominant lineage with BCR-ABL+ cells detectable by qRT-PCR but no CFCs were identified. Secondary recipients of these cells were reconstituted with myeloid cells but only transiently. Conclusion Functionally defined chronic phase CML stem cells represent a very minor subset of the CD34+CD38- compartment and assessment of these cells, like assessment of the most commonly used 5-week LTC-IC assay, overestimates the BCR-ABL+ stem cell compartment. Both of these endpoints also fail to provide a reliable indicator of the prevalence of BCR-ABL+ stem cells defined by more stringent functional assays. We also show for the first time that BCR-ABL+ CML stem cells are capable of serial transplantability spanning one and a half years in NSG mice and this may be anticipated by results from the 6-week LTC-IC assay. Disclosures: No relevant conflicts of interest to declare.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.