TKI Selection in CML Demands Nuanced Approaches

Choosing the most effective TKIs for individual patients with chronic myeloid leukemia (CML) requires decision making based on efficacy data, agent safety profiles, and patient characteristics, according to Michael Deininger, MD, who gave a presentation on TKI selection during the 11th Annual Society of Hematologic Oncology (SOHO 2023) meeting.¹

“A practical approach is that we should always consider what other decision-making criteria affect us that influence this decision, and these are treatment goals, disease risk, costs, and comorbid conditions,” Deininger, chief scientific officer, Versiti Blood Research Institute, said during a presentation of the results at SOHO 2023. “Depending on where the patient is in their journey with CML, they may weigh these criteria in different ways.”

There are several options the frontline setting for patients with chronic phase CML, with major randomized phase 3 trials showing long-term results for imatinib (Gleevec; Novartis), nilotinib (Tasigna; Novartis), dasatinib (Sprycel; Bristol Myers Squibb), and bosutinib (Bosulif; Pfizer). As a first step, Deininger started by looking at the clinical efficacy for each of these agents in the frontline setting.

In totality, the studies were more similar than different. Regardless of the frontline TKI selected, the overall survival (OS) is now 90% or more at 60 months. Across the studies, there have been higher major molecular responses (MMR) observed with the second-generation TKIs, namely nilotinib, dasatinib, and bosutinib; however, this did not translate into better OS with these agents, said Deininger. “For some reason, probably through effective salvage, [response] does not translate into increased survival,” he said.

Additionally, higher rates of deep molecular response (MR4.5) were seen with the second-generation TKIs, although this end point was not reported with imatinib. Of these second-generation agents, treatment-free remission (TFR) has only been reported in a phase 3 study for nilotinib.

In a 10-year analysis of the phase 3 ENESTnd study (NCT00471497),² the estimated eligibility for TFR was 48.6% with a 300 mg twice daily dose of nilotinib. With the 400 mg twice daily dose it was 47.3% and with imatinib it was 29.7%. The EURO-SKI trial (NCT01596114)³ also examined TFR across TKIs and found a molecular recurrence-free survival (MRecFS) rate of 48% and a molecular recurrence- and treatment-free survival (MRecTFS) at 36 months of 46%. “If you include both first, second, and third attempt to achieve TFR, then probably about 30% of patients will be able to [remain in remission when treatment free],” Deininger said.

The discussion of whether the treatment goal will be TFR or OS typically takes place at the 12-month mark. “If survival is the goal, you may be happy with a complete cytogenetic response but if TFR is the goal you may think about changing the regimen,” Deininger said. With survival as the goal, treatment may not be stopped, which creates the potential for more adverse events.

To explore the potential risks associated with treatment, Deininger turned to the adverse events seen in the ENESTnd study,² particularly the cardiovascular (CV) events. By the 10-year mark, there was a higher cumulative rate of CV events seen with nilotinib vs imatinib. At the 300 mg dose 16.5% reported a CV event and at the 400 mg dose this was 23.5%. In the imatinib arm, the CV event rate was 3.6%.

“These are the results for the ENESTnd trial, but I think they apply to other second-generation TKIs just as much,” Deininger said. “The event rate is a lot higher with nilotinib and it doesn't seem to go away very easy or quickly. There is an accumulation over time that needs to be taken into consideration. By and large, all of the second-generation TKIs do increase cardiovascular risk.”

Before administering treatment, past medical history should be taken into consideration, particularly a history of CV events like chronic heart failure, QT prolongation, or impaired left fraction. “There are no real absolute contraindications. If the disease risk mandates a more potent agent, then this is still the driving factor. If you are not working under this constraint, you can pick the TKI based on comorbidities,” said Deininger.

Another method for avoiding AEs might be dose selection. In a 2022 propensity score analysis,⁴ low dose dasatinib at 50 mg per day was compared with standard dosing at 100 mg per day as a frontline therapy for chronic phase CML. With 60 months of follow up, the 3-year MR4.5 rates were 77% with low-dose dasatinib and 62% with the standard dose (P = .02). OS rates were 97% with the low dose vs 96% with standard dosing. The rate of pleural effusion was 5% with the low dose compared with 21% with standard dose.

“The dose of these drugs is probably reducible without compromising efficacy, or in fact, even improving efficacy,” said Deininger. “There was not a typical dose-limiting toxicity to follow. Second-generation TKIs were used at relatively high doses based on pharmacokinetic evaluations.”

After the frontline treatment setting, treatment selection in the first salvage setting becomes a more important consideration. There are 3 scenarios for advanced CML, Deininger noted, namely primary or acquired TKI resistance, accelerated or blast phase at diagnosis, or progression to accelerated or blast phase while on a TKI.

In the case of acquired resistance, BCR-ABL kinase domain mutation analysis allows for a rational choice of TKI selection, based on the acquired alteration. As an example, Deininger called attention to T315I, which denotes resistance to most second-generation TKIs. For this alteration, however, the third generation TKI ponatinib (Iclusig; Takeda) has shown moderate efficacy. “At this time, ponatinib, as a third-generation inhibitor is probably the agent of choice if a second-generation agent fails,” he said.

The main concern with ponatinib has been related to CV toxicity, Deininger pointed out. The OPTIC trial (NCT02467270) was undertaken to explore a dose optimization strategy for the agent. In this study,⁵ patients started at a dose of 45 mg, 30 mg, or 15 mg daily. If BCR-ABL1 was reduced to 1% or less, then the dose was reduced to 15 mg. Overall, better responses were seen in the 45 mg and 30 mg starting dose groups, and the 45 mg reduced to 15 mg was shown to be the most ideal dose. Findings varied by starting BCR-ABL1 levels and by type of mutation, offering other guides for dose selection.

In prior studies, Deininger noted that CV events with ponatinib were roughly 15% to 28%, whereas in the OPTIC trial, the grade 3/4 adjudicated treatment-emergent arterial occlusive events were 4% at the lowest dose to 6% in both the 45 mg and 30 mg starting arms. There were no grade 5 occlusive events reported.

The traditional TKIs bind to the ATP binding pocket; however, a newer agent, asciminib (Scemblix; Novartis) binds to the myristoyl pocket. The ASCEMBL trial (NCT03106779) compared asciminib to bosutinib following 2 or more prior TKIs. The trial excluded patients with T315I or V299L mutations, given bosutinib’s lack of efficacy in these groups. At 24 weeks, the MR4.5 rate was 8.9% with asciminib vs 1.3% with bosutinib. MMR at 24 weeks was achieved for 25.5% of those treated with asciminib vs 13.2% of those in the bosutinib arm.

“In terms of toxicity, it turns out that asciminib is well tolerated, and probably better tolerated than bosutinib,” Deininger said.

In the United States and other developed countries, accelerated or blast phase is uncommon at diagnosis. In those with accelerated phase, the disease is largely treatable with TKIs, with decent outcomes, Deininger said. Those diagnosed with blast phase CML are more difficult to treat, and often cannot be salvaged with a TKI. In these individuals, allogeneic stem cell transplant remains the standard.

In terms of the use of allogeneic transplant, Deininger felt it was no longer appropriate for most patients with chronic phase CML, except in the third line after other options have been exhausted. For those with advanced phase disease, he recommended a TKI first and would consider transplant upon progression. For blast phase at diagnosis, allogeneic transplant could be used throughout the treatment landscape.

References

1. Deininger, Michael. Rational Choice of TKI. Presented at: 11th Annual Meeting of the Society Hematologic Oncology (SOHO 2023), September 7, 2023. Houston, TX.
2. Kantarjian HM, Hughes TP, Larson RA, et al. Long-term outcomes with frontline nilotinib versus imatinib in newly diagnosed chronic myeloid leukemia in chronic phase: ENESTnd 10-year analysis. Leukemia. 2021;35(2):440-453. doi: 10.1038/s41375-020-01111-2.
3. Mahon F-X, Richter J, Hochhaus A, et al. FINAL Analysis of a PAN European STOP Tyrosine Kinase Inhibitor Trial in Chronic Myeloid Leukemia: The EURO-SKI Study. Blood. 2021;138 (Suppl 1):633. doi:10.1182/blood-2021-148369
4. Jabbour E, Sasaki K, Haddad FG, et al. Low-dose dasatinib 50 mg/day versus standard-dose dasatinib 100 mg/day as frontline therapy in chronic myeloid leukemia in chronic phase: A propensity score analysis. Am J Hematol. 2022;97(11):1413-1418. doi: 10.1002/ajh.26689.
5. Cortes JE, Deininger MW, Lomaia E, et al. Three-Year Update from the Optic Trial: A Dose-Optimization Study of 3 Starting Doses of Ponatinib. Blood. 2022;140(Suppl 1):1495-1497.doi:10.1182/blood-2022-157822.
6. Réa D, Mauro MJ, Boquimpani C, et al. A phase 3, open-label, randomized study of asciminib, a STAMP inhibitor, vs bosutinib in CML after 2 or more prior TKIs. Blood. 2021;138(21):2031-2041.