Background: Progression and relapse in Multiple Myeloma (MM) is induced by changes in the clonal tumor composition. In order to better understand the mechanisms underlying these dynamics, we developed clonal competition models based on the co-culture of fluorescent labelled isogenic MM cells, with or without the alteration under study.

Methods: To understand the effect of mono- and bi-allelic TP53 lesions, we use the AMO1 cell line, one of few myeloma cell lines harbouring wild type TP53 (WT). After modification with CRISPR / CAS9, we selected subclones with mono- and/or bi-allelic deletion of TP53. For the characterization of alterations in RAS, we selected OPM2 cells, one of the few lines with the RAS pathway intact. Furthermore, we generated the KRAS WT, G12A and A146T sublines by stable transfection with Sleeping Beauty vectors. To study mutations related to resistance to IMIDs and PIs, we introduced mutations in the target genes IKZF1 (WT, A152T, Q170D or R439H), CUL4B (KO), and PSMB5 (WT or A20T ) in AMO1 and L363, cell lines sensitive to IMiD or PI treatment. In addition, all WT and mutant sublines were also stably transformed with E-GFP or LSS-mkate2-RFP for flow cytometry analysis.

Results: We recently demonstrated that lesions in TP53, both mono- and bi-allelic, induce a growth advantage to the affected cells. In the current study, we also observed an increased fitness in KRAS mutated cells (G12A or A146T vs WT) independent of treatment. We co-cultivated KRAS mutant with WT cells at a ratio 1:3 in two independent experiments, with the color labelling switched (red/green wt/mutant and vice-versa). KRAS G12A clone significantly expanded and reached 50% of the cells at day 40. Likewise, A146T clone outcompeted WT cells, but the time required to represent the majority of cells in the coculture was longer.

We next explored the effects of resistance mutations and drug exposure. Both the IKZF1 A152T and CUL4B KO mutants outcompete WT cells in the presence of Lenalidomide (LEN). The same effect was observed for the PSMB5 A20T mutant exposed to Bortezomib (BOR). This selection ("Survival Fitness") did not occur without the presence of the drug. Thus, resistance related mutations seem only to provide a fitness advantage under drug exposition. In addition, both the CUL4B KO and PSMB5 A20T mutants were overcome by WT cells when the drug was removed from co-culture, suggesting that these lesions provide a survival disadvantage without the selective pressure of IMiD or PI. This may provide an explanation for the low mutation rate in this gene in recent sequencing publications, as usually samples are not obtained under selective pressure but in treatment free intervals. IKZF1 mutations outside the IMiDs / CRBN binding area (Q170D and R439H) provided no advantage to the cells.

Conclusions: Our clonal competition assays provide novel insights on the impact of point mutations on the fitness of affected myeloma subclones, either with or without the selective pressure of therapy.

Disclosures

Martinez-Lopez:Celgene: Honoraria, Other: Advisory boards and Non-Financial Support ; Amgen: Honoraria, Other: Non-Financial Support ; F. Hoffmann-La Roche Ltd: Honoraria; Janssen: Honoraria, Other: Advisory boards and Non-Financial Support ; BMS: Honoraria, Other: Advisory boards; Incyte: Honoraria, Other: Advisory boards; Novartis: Honoraria, Other: Advisory boards; VIVIA Biotech: Honoraria.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution