The amount of descriptive genetic data in multiple myeloma (MM) is growing exponentially. Clonal dynamics leading to disease progression and resistance development are evident and have been described in a number of exemplary MM patients. Still, for most identified genetic lesions, the specific impact on clinic and biology remains unclear.
Aiming for a better understanding of the underlying mechanisms of clonal progression, we developed clonal competition models, based on co-culture of fluorescence-marked isogenic MM cells with or without the alteration under study. This experimental setting allows us to study the implications of a specific lesion and its interactions with different environmental conditions such as drug exposition.
To study the effect of mono- and bi-allelic TP53 lesions and the impact of proteasome subunit mutations we selected the TP53 wild-type (WT) cell line AMO1. As this cell line has a weak response to IMiDs, we generated L363 sublines with IKZF1 or PSMB5 alterations and MM1S cells with IKZF1 or TP53 lesions. All these sublines and their parental (WT) controls were then stably marked with EGFP or LSS mKate RFP. Then, the co-cultures with mutant and WT cells were monitored by flow cytometry under different therapeutic conditions.
Exploring the CoMMpass dataset, we detected one patient (MMRF_1152) who acquired two sub-clonal nonsense CUL4B mutations after IMiD exposition. In our clonal competition assays, CUL4B KO cells were selected in the presence of LEN but they induced a survival disadvantage when the drug was not present (Figure, top). We previously described a similar observation regarding PSMB5 mutations affecting the chymotrypsin-like catalytic core (PI binding). In vitro these alterations generated an impasse on the proteasome activity that potentially affects cell growth (Barrio S, et al. Leukemia 2019). Our results suggest that some of the acquired mutations may induce survival fitness under drug exposition but represent a disadvantage when the therapy is removed. This could explain why such alterations are not being detected more frequently in relapsed MM patients. Of note, however, a hotspot mutation identified in the LEN binding area of IKZF1 (A152T) did not have any negative impact on cell growth without treatment but was selected in the presence of LEN. IKZF1 mutations outside the LEN binding area (Q170D and R439H) did not provide any advantage to the cells. Besides survival fitness, we have also observed growth fitness, a selection that occurs independently of therapy. When cells with mono- or bi-allelic TP53 lesions were co-cultured with WT cells, a strong growth dynamic was observed. This advantage remained under exposure to Melphalan (MEL) or Bortezomib. Of interest, MEL seems to increase the fitness of cells with mono-allelic TP53 alterations.
In summary, we observed three different types of fitness advantages (Figure, bottom): Negative survival fitness, induced by drug exposition but with negative impact on basal cell growth as described for CUL4B or PSMB5. Neutral survival fitness (IKZF1), without this negative effect when the drug is absent. Growth fitness, independent of treatment exposition (TP53 mono- and bi-allelic lesions). Altogether, our data suggest that it is possible to apply clonal competition assays to perform analytical or even quantitative genetics. This approach might help in the future to select the best regimen to control the fitness advantages induced in MM cells by patient-specific lesions.