Determination of baseline inhibitory concentration of ASC and ABPIs monotherapy in wild-type CML cell lines
The study first determined the baseline inhibitory drug concentration for ASC monotherapy and each ABPI using the K562/WT cell line and BaF3/WT cell lines, respectively (Fig. 1). In the K562/WT cell line, the cell inhibition rates were measured as follows: IMA (136 nM) − 59.8%, BOS (6 nM) − 56.9%, DAS (0.3 nM) − 61.5%, NIL (16nM) − 59.9%, PON (0.4 nM) − 78.6%, ASC (6 nM) − 71.6–79.4%. In the BaF3/WT cell line, the cell inhibition rates were measured as follows: IMA (150 nM) − 79.6%, BOS (10 nM) – 57.5%, DAS (0.8 nM) − 72.5%, NIL (10 nM) − 74.8%, PON (0.6 nM) − 77.3%, ASC (3 nM) − 47–70.5%. These drug concentrations for each drug were used as the baseline concentration for the next step.
Determination of baseline inhibitory concentration for combination therapy of ASC with ABPIs in wild type CML cell lines
The combined treatment of ASC with ABPIs at the baseline inhibitory concentration showed 100% cell inhibition in both BaF3/WT and K562/WT cell lines. The combination of ASC with ABPIs at half of the baseline inhibitory concentration continued to exhibit a higher inhibitory effect than single TKI treatments in both cell lines. In the K562 cell line, treatment with IMA at 136nM alone showed an inhibitory effect of 59.8%, while the combined treatment of half of the baseline inhibitory concentration of IMA (68nM) and ASC at 3 nM showed a 4% higher inhibitory effect of 63.7% (Fig. 1A). Similarly, the combination of ASC at 3 nM with half of baseline inhibitory concentrations of DAS, NIL, and BOS showed 6.1%, 10.8%, and 6.6% higher inhibitory effects compared to TKI treatment at the baseline inhibitory concentration alone. Of interest, combined treatment with ASC at 6 nM with half of the baseline inhibitory concentration of ABPIs showed a significant increase in inhibitory activity to 89.5–96.9%, compared with ABPI treatment alone at the baseline inhibitory concentration. Likewise, in the BaF3/WT cell line, the combination of ASC at 3 nM with the half of the baseline inhibitory concentration of ABPIs exhibited a higher inhibitory efficiency than ABPIs alone at their baseline inhibitory concentration (Fig. 1B). This finding suggests that the combination of ASC with a reduced dose of ABPIs can feasibly and effectively inhibit wild-type CML cells.
Different drug sensitivity of asciminib monotherapy in BaF3 CML cell lines according to the ABL1 kinase domain mutation profile
While it is well established that different types of ABL1 KD mutant CML exhibit varying susceptibilities to different ABPIs, the resistance profile of ABL1-KDM to ASC has not been fully established yet, despite rapidly growing evidence. Using the data from the BaF3 cell line results, we normalized the inhibitory effect of various BaF3 cell lines using the cell inhibition percentage with 3 nM of ASC on BaF3/WT as the baseline (Fig. 2) (Supplementary Table. 1).
The G250E, E255K and T315A mutations were indeed highly susceptible to ASC monotherapy, exhibiting 1.29–2.12 times higher cell inhibition percentages. These cell lines showed a 0.08-0.85-fold lower inhibition rate when treated with first- and second-generation ABPI alone. When combined with ASC 3 nM, there was a 1.86-2.13-fold higher inhibition rate. Thus, we have classified this group as a highly sensitive mutations to both ASC monotherapy and combination therapy.
The second group consists of the M351T and F317L mutations, which show resistance to ASC monotherapy at 3 nM but could be overcome with dose escalation to 6nM or by adding other ABPIs. In these cell lines, ASC at 3 nM showed inhibition compared to BaF3/WT at 3 nM ASC treatment, but dose escalation to 6nM resulted in improved inhibition to 1.3-1.36-fold. Regardless of the type of ABPI, combination therapy with ASC at 3nM showed a 1.35 to 1.65-fold improved inhibitory effect. Accordingly, this group can be classified as intermediate sensitivity.
The last group includes F317V, H396P, Y253F, M244V, and T315I. This group could not be inhibited even with doubled concentration of ASC. Regarding sensitivity to the combination of ASC with ABPIs, inhibitory activity varies and depends on which ABPI is combined.
For instance, when combined with ASC 3 nM, the F317V mutation showed 1.45-fold and 1.67-fold higher inhibitory effects with IMA (150nM) and NIL (10nM), while the addition of BOS (10nM) and DAS (0.8nM) resulted in reduced inhibitory activity to 0.68 and 0.39 fold, respectively. Similarly, when combined with ASC 3 nM, H396P and Y253F showed 1.32-1.57-fold higher inhibitory effects with BOS (10nM) or DAS (0.8nM). However, M244V and T315I failed to restore their sensitivity even with combinations of ABPIs with escalated doses of ASC at 6nM. Consequently, overcoming resistance in the case of M244V and T315I mutations will require much higher concentration of ASC.
Additionally, the combination of ASC 3 nM with doubled dose of third-generation PON (1.2nM from the baseline concentration) could overcome resistance, showing 1.70-fold and 1.28-fold higher inhibitory effects to each mutation. We have not tested combinations of a higher concentration of ASC over 6nM with other ABPIs in M244V or T315I mutant BaF3 cell lines.
In summary, we have classified the BaF3 cell lines into 3 categories based on their sensitivity patterns to ASC doses and the combination of ASC with other ABPIs:
1) Highly sensitive (G250E, E255K, T315A): ASC 3nM monotherapy results in over 50% cancer cell death. This group is also highly sensitive to the combination of ASC and ABPI, to the extent that a combination of 3 nM ASC with ABPI at half the baseline concentration can overcome resistance by 1.54-2.12-fold.
2) Intermediate sensitivity (M351T, F317L): ASC 3nM monotherapy could not achieve sufficient inhibition, but an increased dose of ASC to 6nM could overcome its resistance. Additionally, the combination of ASC and ABPI at baseline concentration can overcome resistance with a 1.35-1.66-fold inhibitory effect.
3) Low sensitivity (F317V, H396P, Y253F, M244V, and T315I): In the case of ASC 6nM monotherapy, it does not have an inhibitory effect. The cell killing effect can be restored when ASC is combined with a specific ABPI. When considering combinations with ASC, the combination of IMA and NIL is good for F317V, and BOS and DAS are effective for H396P and Y253F.
These results suggest that dual blockade is more effective for specific subtypes of ABL1 KDM, and that a reduced dose of ABPI is sufficient to inhibit ABL1 KD mutant BaF3 cell lines when combined with ASC.
Efficacy of ponatinib combined with asciminib according to mutation profiles of BaF3 CML cell lines
In most of the BaF3 cell lines with different ABL KDMs, the combined treatment of ASC at 3 nM and PON at 0.3 nM showed similar activity to that of PON at 0.6 nM treatment alone. For instance, in the case of T315A, M351T, and F317V, the combination of ASC 3 nM and PON 0.3 nM exhibited approximately 30% higher inhibitory activity than PON at 0.6 nM alone. In F317L, H396P, and Y253F mutant cell lines, the combination of ASC at 3 nM with PON at 0.6 nM exhibited 1.6-1.8-fold higher efficacy than ASC at 3 nM monotherapy. However, for E255K, M244V, and T315I, the combination of ASC at 3 nM with PON did not enhance inhibitory activity, even when the dose of PON was doubled to 1.2 nM.
Determination of ABPI dose in combination with fixed-dose asciminib against ABL1 kinase domain mutant CML cell lines
Initially, we hypothesized that a combination of a fixed dose of ASC with a reduced dose of ABPI could effectively inhibit ABL1-KDM mutant CML cell lines. Various doses of ABPIs were tested in combination with ASC, evaluating their efficacy across all subtypes of ABL1 KD mutations. Overall, when combined with ASC at 3 nM, DAS and PON showed significantly enhanced inhibition, even at half the concentration of DAS and PON from the baseline concentration (p < 0.05(*), p < 0.01(**), respectively)(Fig. 3). Interestingly, the combination of PON at 0.15 nM and ASC at 3 nM exhibited higher inhibitory activity (%) than PON monotherapy at 0.3 nM, while it showed a similar inhibition to PON monotherapy at 0.6 nM. This suggests that the PON dose can be reduced to a quarter of the baseline concentration (i.e. 0.15 nM) when combined with ASC, while maintaining similar efficacy to full-dose PON therapy (i.e. 0.6 nM).
Evaluation of synergistic inhibitory activity of dual blockade of asciminib with ABPIs
To assess the synergistic inhibitory activity of dual blockade using ASC and ABPIs, ZIP synergy scores and responses were calculated (Supplementary Fig. 1). In Fig. 4, significant synergistic effects were indicated by ZIP score of 10 or above, represented by a darker brown color, while the response (i.e., therapeutic efficacy) was shown by the radius of each circle. The results of the synergy score are summarized in Fig. 4. While some combinations exhibited enhanced inhibition with dual blockade, it was determined to be merely additive, not synergistic. Overall, the best combination of ASC with DAS, except for F317V/T315I, or ASC with PON, appeared to be best combination considering both inhibitory efficacy and synergism.
In BaF3/WT cells, combining ASC with all types of ABPIs demonstrated high sensitivity. For the G250E and E255K mutations, the combination of ASC at 3 nM and ABPIs resulted in a high inhibitory effect with a 100% inhibition rate, making it impossible to calculate the synergy score for these combinations. However, in the case of the T315A mutation, the combination of ASC with all ABPIs increased the inhibition by approximately 1.9-fold compared to ASC at 3 nM in BaF3/WT cells, with significant synergistic effects (ZIP score = 10.41–25.74).
Moreover, the M351T, H396P, and Y253F mutations also exhibited an enhanced synergistic effect when combined with DAS at 0.8nM and ASC at 3 nM, resulting in synergy scores of 16.26, 14.27, and 45.72, respectively. While F317L showed an enhanced synergistic effect when combined DAS at 0.8nM and ASC at 3 nM (ZIP score = 19.97), F317V did not display significant synergistic effects from this combination (ZIP score = 0.25). However, the combination of ASC with NIL exhibited a synergistic inhibitory effect against F317V (ZIP score = 22.33).
Regarding H396P, co-treatment of ASC with DAS or BOS, rather than with IMA or NIL, exhibited higher synergistic inhibition: ZIP synergy score were 14.27 with DAS, 22.07 with BOS, while they were 10.51 with IMA, and 9.44 with NIL. In the case of PON, a notable synergy effect was observed. The combination treatment of PON at 0.3nM with ASC at 3 nM exhibited a synergistic effect across all the KDMs except E255K (ZIP score = 3)(Supplementary Fig. 1). Although the synergistic score was quite high, M244V and T315I mutations exhibited similar or slightly lower inhibition (0.68–1.02 fold) compared to WT when treated with ASC monotherapy at 3 nM, suggesting that a higher concentration of ASC in combination with PON will be required to overcome M244V and T315I mutations.
To confirm our hypothesis that dual blockade of the ABL1 kinase protein can lower the required dose of PON, we assessed the pCrkL/CrkL ratio, which can provide direct evidence of BCR-ABL1 protein inhibition (Fig. 5)(ref. 30, 31). Surprisingly, a lower concentration of PON, half of the baseline inhibition concentration, combined with 6 nM ASC significantly reduced BCR-ABL1 protein activity compared to dose escalation of PON from 0.6 nM to 1.2 nM (p < 0.001). Additionally, BCR-ABL1 activity was significantly suppressed when 3 nM ASC was added to the baseline inhibition concentration of PON (i.e. 0.6 nM), compared to 0.6 nM PON alone (p < 0.01).