Multiple TKIs for first-line CML treatment ‘a mixed blessing’

The treatment landscape for most blood cancers has evolved rapidly during the past decade, and chronic myeloid leukemia is no exception.

Imatinib (Gleevec, Novartis) became the de facto gold standard in 2001 after it became the first FDA-approved tyrosine kinase inhibitor indicated for the treatment of CML.

Subsequent FDA approvals of two other TKIs — dasatinib (Sprycel, Bristol-Myers Squibb) in 2006 and nilotinib (Tasigna, Novartis) in 2007 — further expanded clinicians’ armamentarium, helping to transform CML from a lethal disease to a chronic condition.

Yet, they also created what an ASH education program report published last year labeled “an embarrassment of riches.” Despite greater response seen with nilotinib and dasatinib, neither has demonstrated a clear survival advantage over imatinib, and the decision about which first-line therapy is appropriate for which patients continues to vex treatment teams.

“It’s unusual to have a cancer diagnosis with three approved first-line therapies that are so strong,” Michael J. Mauro, MD, leader of the Myeloproliferative Neoplasms Program at Memorial Sloan Kettering Cancer Center and HemOnc Today’s leukemia section editor, told HemOnc Today. “Adding all factors together that go into determining which TKI is best is a tremendous challenge. It requires intimate knowledge of each drug, including the subtle differences with regard to response and side effect profiles.”

Clinicians also must monitor patients closely, as a prompt switch to another TKI after first-line treatment failure provides the greatest chance for meaningful response.

“As part of this process, of course, one must have an understanding of the disease biology,” Mauro said. “Making subtle adjustments to stay ahead of the leukemia is imperative.”

HemOnc Today spoke with several CML experts about the factors that influence their choice of first-line treatment, the importance of patient monitoring, the factors that influence their decision to switch treatment regimens, and the possibility that treatment decisions could become more complex as a result of issues such as reimbursement, toxicity or long-term trial results.

A subjective assessment

No single factor determines which patients should receive which TKIs for first-line treatment of chronic-phase CML. A patient’s age, disease stage, comorbidities and risk profile all must be taken into account, but uncertainty always remains.

Mauro emphasized the need for additional data to help clarify the ideal settings for each approved first-line therapy, as well as long-term follow-up of studies designed to assess the efficacy of additional second- and third-generation TKIs, such as bosutinib (Bosulif, Pfizer) and ponatinib (Iclusig, Ariad).

Each TKI offers differences in schedule, common manageable side effects and severe adverse-event profiles, Mauro said. Certain patients may be more susceptible to specific adverse events, but there is no consensus on potential contraindications.

“We are really left with a fairly subjective assessment of multiple factors rather than a data set that can allow us to develop a clear algorithm or clear pathway that dictates which drugs should be used in which situations,” Mauro said.

Because imatinib has been an established first-line treatment for more than a decade, most clinicians understand its potential for adverse events and the appropriate strategies to manage them. In some cases, that knowledge could push the decision about first-line treatment choice in imatinib’s favor.

“That being said, we have seen extensive research and good science that support the fact that dasatinib and nilotinib offer advantages over imatinib with regard to depth of response and rapidity of response,” Mauro said. “At the same time, trials haven’t shown significant differences in overall outcome in terms of survival, and it’s uncertain as to why this is.”

The results of the ENESTnd and DASISION trials — both of which were published in 2010 in The New England Journal of Medicine — convinced most oncologists that dasatinib and nilotinib were effective alternatives to imatinib for patients with newly diagnosed chronic-phase CML.

The ENESTnd trial showed twice-daily nilotinib was superior to imatinib in patients with newly diagnosed, chronic-phase Philadelphia chromosome-positive CML. In the DASISON trial, researchers observed significantly higher and faster rates of major molecular response and cytogenetic response among patients assigned once-daily dasatinib compared with those assigned once-daily imatinib.

A subsequent review by Pavey and colleagues, published in 2012 in Health Technology Assessment, was designed to compare the clinical effectiveness and cost-effectiveness of imatinib, dasatinib and nilotinib.

In the comparison of dasatinib vs. imatinib, researchers reported significantly higher rates of complete cytogenetic response (83% vs. 72%) and major molecular response (46% vs. 28%) at 1 year in dasatinib-treated patients (P<.0001 for both). The comparison of nilotinib vs. imatinib showed significantly higher rates of complete cytogenetic response (80% vs. 65%; P<.001) and major molecular response (44% vs. 22%; P<.0001) at 1 year in nilotinib-treated patients.

However, 48- to 60-month follow-up of both comparisons showed no statistically significant difference in survival between imatinib and either of the other two agents.

Although imatinib was the first approved TKI for CML and is the subject of a larger body of evidence, the second-generation inhibitors are more potent, according to Michael W. Deininger, MD, PhD, professor and chief of the division of hematology and hematologic malignancies at Huntsman Cancer Institute at the University of Utah.

At the same time, longer follow-up has revealed side effects that must be taken into account.

“We have a scenario in which we have one very established agent — imatinib — with an excellent safety profile, but it isn’t a very potent inhibitor,” Deininger told HemOnc Today. “Conversely, the other two drugs haven’t been around as long, we have less experience with them and they are not quite as benign in their safety profiles, but they do have the advantage of being more potent. The critical question is whether we can identify factors that will sway us in one direction or another in any given patient.”

Long-term monitoring

Despite the effectiveness of all three approved first-line agents and the potential for treatment-free remission, frequent monitoring of patients with chronic-phase CML is essential.

Long-term molecular monitoring identifies complete molecular responders, detects relapse early and, thus, may help prevent transformation to blast crisis, which is associated with rapid progression and short survival.

Monitoring also helps ensure treatment compliance, Deininger said.

“Patients will sometimes take ‘drug holidays,’ and this becomes apparent with polymerase chain reaction (PCR) monitoring and is a good way to keep patients in line with treatment adherence,” he said. “The bottom line is, it is important to properly monitor all CML patients because it helps us to exploit the great efficacy that these drugs have for CML.”

Two guidelines on CML management — those from the National Comprehensive Cancer Network and the European LeukemiaNet (ELN) — offer similar recommendations for long-term monitoring.

The NCCN recommends quantitative PCR every 3 months for 3 years after a complete cytogenetic response, followed by every 3 to 6 months thereafter.

ELN recommends qualitative PCR every 3 months until a major molecular response is achieved, then every 3 to 6 months thereafter. ELN also recommends quantitative PCR be repeated in 1 to 3 months if there has been a 1-log — or a five- to 10-fold — increase in BCR-ABL1 transcripts compared with the previous value.

An estimated 15% to 20% of patients with chronic-phase CML are not monitored sufficiently to assess whether they are responding to treatment, according to Elias Jabbour, MD, associate professor in the department of leukemia in the division of cancer medicine at The University of Texas MD Anderson Cancer Center.

“This is really bad,” Jabbour said. “Multiple studies suggest outcome depends upon how the patient responds early on. It’s important for the patient to not only take their medication and for physicians to optimize safety concerns, but to also regularly monitor these patients.”

This monitoring plays a vital role in determining whether a new treatment strategy must be considered, according to Timothy P. Hughes, MD, and Deborah L. White, PhD, both of South Australian Health and Medical Research Institute and University of Adelaide in Australia.

“Whatever first choice of TKI is made, treatment failure or intolerance must be recognized early because a prompt switch to another TKI likely provides the best chance of achieving optimal response,” Hughes and White wrote in a paper published as part of an ASH education program in 2013.

In a separate paper published in Blood in July, Hughes and colleagues found patients with chronic-phase CML who had persistent minimal residual disease after long-term treatment with imatinib achieved deeper molecular responses and undetectable disease when they switched to treatment with nilotinib. However, the improved response came at the expense of increased toxicities.

“In patients with molecular responses less than MR^4.5, the clinical benefits of switching to nilotinib to achieve deeper molecular responses should be carefully weighed against the potential for adverse events, including cardiovascular adverse events, on an individual basis,” Hughes and colleagues wrote. “Any adverse event of any grade that requires a switch in therapy may be considered a competing risk event for the achievement of response on that TKI. … Adverse events that occur after TKI switch must be viewed in light of the fact that patients who achieve deeper responses such as MR^4.5 are likely to have better long-term outcomes.”

Because there are so many treatment options, clinicians sometimes switch therapies too soon, according to Jerald P. Radich, MD, a member of the Clinical Research Division at Fred Hutchinson Cancer Research Center, professor in the division of oncology at University of Washington School of Medicine in Seattle, and a member of both the NCCN and ELN guideline committees.

“Patients are often diagnosed with CML circumstantially. They do not feel sick, so to speak,” Radich said. “If there are any side effects [of first-line treatment], patients are obviously disturbed by this, and there is always the pressure to switch treatment quickly. In most cases, though, the toxicities would greatly decrease with time and patience. It’s the same thing with response: If someone isn’t responding right away, they often are switched, but had they been a little more patient they would have been fine. In some ways, it’s a mixed blessing to have so many treatment options.”

Cardiovascular comorbidities

Cardiovascular health is another essential component of patient monitoring.

“New side effects that have appeared with more use of dasatinib and nilotinib may lead to treatment interruptions and morbidity/mortality, and these issues are under very close scrutiny,” Mauro said. “Anything that detracts from the success of a CML patient is important because the overall expectations are so good. Patients today are very likely to achieve a protective and deep remission that is going to be long lasting. Serious adverse events can deteriorate such a bright outlook.”

In the Blood paper by Hughes and colleagues, patients in the nilotinib arm experienced a higher rate of grade 3 or grade 4 adverse events (48.5% vs. 22.3%) and more frequently discontinued therapy due to an adverse event of any grade (19.8% vs. 5.8%). Ischemic heart disease and peripheral artery occlusive disease were infrequent, but more cases were reported in the nilotinib arm than the imatinib arm. All three patients with peripheral artery occlusive disease had pre-existing cardiovascular disease risk factors.

Results from a study by Kim and colleagues, published in 2013 in Leukemia, suggested that patients who were either currently undergoing treatment with nilotinib or were previously exposed to nilotinib were at greater risk for peripheral artery occlusive disease than those treated with imatinib.

Kim and colleagues evaluated clinical and biochemical risk factors for cardiovascular disease in 159 patients with chronic-phase CML. They calculated a relative risk for peripheral artery occlusive disease of 10.3 among patients who received first-line nilotinib vs. those who received first-line imatinib. Researchers concluded noninvasive monitoring and careful assessment of patients’ risk factors is warranted.

“There are more data in the literature now about the cardiovascular adverse effects of nilotinib, in particular peripheral arterial occlusive disease,” Deininger said. “For dasatinib, there is increased risk for pulmonary hypertension, and this needs to be a part of the discussion with patients. It depends on the therapeutic goals whether these potential side effects are acceptable.”

BCR-ABL mutations

The 5-year survival rate for CML now stands at 95% vs. 30% before the advent of TKIs. However, an estimated 20% to 30% of patients become resistant to one or more agents.

Deininger and colleagues examined BCR-ABL compound mutants containing two mutations in the same molecule that appear to render some or all TKIs ineffective.

“Our study looked at resistance to the third-line TKI ponatinib,” Deininger said. “Some patients with advanced CML exhibit compound mutations in BCR-ABL, the driving genetic lesion for TKIs. These compound mutations confer a high level of resistance to TKIs, with some differences according to mutation type.”

Additional findings of the study, published in August in Cancer Cell, suggested the compound mutations — in which a T315I mutation is one of the components — typically is resistant to all currently approved TKIs.

Patients with these mutations do not have a TKI option available, and a need clearly exists to develop a treatment that can target these mutations, Deininger said.

“Fortunately, the problems we are studying affect a minority of CML patients, but still leaves some patients with no good treatment option,” researcher Thomas J. O’Hare, PhD, an associate professor of medicine in the division of hematology and hematologic malignancies at University of Utah, said in a press release. “Our goal is to have a TKI option for every patient.”

The findings in CML will provide a blueprint for contending with resistance in other highly aggressive diseases such as non–small cell lung cancer and acute myeloid leukemia, O’Hare said.

Treatment cessation

Researchers now are investigating the possibility that patients who undergo upfront treatment with second-generation TKIs could eventually discontinue treatment.

“This would change how we treat patients because the goal would then shift toward achieving a complete molecular response and then eventually discontinuation of treatment,” Deininger said. “Based upon this paradigm, the second-generation TKIs induce more rapid and more profound responses, so the hope is more profound responses will enable more patients to go off treatment. This is still something in the realm of clinical trials. If these data can be generated, it will be a pivotal moment in choosing front-line chronic-phase CML treatment.”

A paper published in September in the Journal of Medical Case Reports, suggests such a scenario is possible.

Caocci and colleagues reported on a 64-year-old white man diagnosed with chronic-phase CML in 2005. He underwent treatment with imatinib, but he became intolerant to it after 4 years and switched to nilotinib. After 2 years of nilotinib treatment, he opted to discontinue treatment. Undetectable molecular response continued for 30 months after discontinuation.

“Our present case suggests that nilotinib withdrawal is safe for patients with CML who achieve a stable undetectable molecular response,” Caocci and colleagues wrote. “Our patient was homozygous for killer immunoglobulin-like receptor haplotype A, previously reported to be a promising immunogenetic marker for undetectable molecular response. We recommend additional studies to investigate patient immunogenetic profiles and their potential role in complete response to therapy.”

Additional studies examining treatment cessation are ongoing in Europe. If they are successful, they could dramatically alter treatment objectives and also save patients with CML substantial amounts of treatment-related expenses, Jabbour said.

“For CML, treatment is taken for a lifetime,” Jabbour said. “So if you take a drug that costs $7,000 per month, the number becomes huge over a lifetime.”

More changes ahead

Although the development of TKIs has revolutionized the treatment of CML, more changes lay ahead.

Investigations into the safety and efficacy of bosutinib and ponatinib are intensifying, and both may eventually prove useful in front-line settings.

The randomized, phase 3 BELA study compared bosutinib, an oral SRC-ABL1 TKI registered in many countries as a second- or third-line treatment option, with imatinib in 502 patients with newly diagnosed chronic-phase CML.

Initial results, published in 2012 in Journal of Clinical Oncology, showed no statistically significant difference in complete cytogenic response rate — the study’s primary endpoint — between bosutinib-treated and imatinib-treated patients (70% vs. 68%; P=.601). However, bosutinib was associated with a higher rate of major molecular response at 12 months (41% vs. 27%; P<.001), shorter time to major molecular response and complete cytogenic response (P<.001 for both), a lower rate of on-treatment transformation to accelerated/blast phase disease (2% vs. 4%) and fewer CML-related deaths (3 vs. 8).

Results from the 24-month follow-up, published in September in British Journal of Haematology, continued to show a higher rate of major molecular response, as well as a high probability of EFS, in the bosutinib arm.

“Further studies may establish a place for bosutinib in the front-line setting,” Hughes and White wrote in the 2013 ASH report.

Ponatinib is an effective treatment option for patients with resistance to multiple TKIs and those with T315I mutations. A phase 2 trial at MD Anderson Cancer Center and a large phase 3 study that was halted were designed to investigate the agent’s potential as a first-line treatment in chronic-phase CML. However, data from phase 1 and phase 2 studies have raised concerns about its toxicity profile, particularly an association with an increased number of cardiovascular events.

There is considerable scientific interest in identifying combinations that can help eradicate CML, Deininger said.

“None of these approaches are ready for prime time, but maybe in a couple of years,” he said.

However, the first-line treatment landscape definitely will change when the basic compound patent for Gleevec expires in the United States on July 4.

“We know that imatinib will become a generic next year,” Jabbour said. “We hope the drug will be very cheap, and instead of $7,000 per month it will average $500 per month.”

This could trigger a seismic shift in treatment decisions.

“It is hard to justify doing front-line treatment with an expensive drug, and imatinib may become the physician’s treatment of choice due to price,” Jabbour said. “I’d go with the cheapest drug and if the patient doesn’t respond well, then I’d move on to other drugs. These are the factors that will dictate our behavior — essentially a financial reason.”

Physicians also may face external pressure to choose generic imatinib as first-line treatment, Deininger said.

“I expect more pressure from insurance carriers on physicians to start patients on imatinib rather than a second-generation TKI,” he said. “This would certainly affect current practice patterns, and it is a little unpredictable how exactly this will shake out. Part of the uncertainty is that we do not know how these generic drugs will be priced and how the manufacturers of dasatinib and nilotinib will respond to this challenge.” – by Jennifer Southall

References:

Brümmendorf TH. Br J Haematol. 2014;doi:10.1111/bjh.13108. Caocci G. J Med Case Rep. 2014;8:295.

Cortes J. Blood. 2012;120:1390-1397.

Cortes JE. J Clin Oncol. 2012;30:3486-3492.

Hughes T. Hematology Am Soc Hematol Educ Program. 2013;2013:168-175.

Hughes TP. Blood. 2014;124:729-736.

Jabbour E. Clin Lymphoma Myeloma Leuk. 2013;13:646-656.

Kantarjian H. N Engl J Med. 2010;362:2260-2270.

Kim TD. Leukemia. 2013;27:1316-1321.

Pavey T. Health Technol Assess. 2012;16:1-277.

Price KE. Onco Targets Ther. 2013;6:1111-1118.

Saglio G. N Engl J Med. 2010;362:2251-2259.

Zabriski MS. Cancer Cell. 2014;26:428-442.

For more information:

Michael W. Deininger, MD, PhD, can be reached at University of Utah, Huntsman Cancer Institute, 1950 Circle of Hope, Salt Lake City, UT 84112; email: michael.deininger@hci.utah.edu.

Elias Jabbour, MD, can be reached at The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030; email: ejabbour@mdanderson.org.

Michael J. Mauro, MD, can be reached at Memorial Sloan Kettering Cancer Center, 1275 York Ave., New York, NY 10065; email: maurom@mskcc.org.

Jerald P. Radich, MD, can be reached at Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N, Seattle, WA 98109; email: jradich@fhcrc.org.

Disclosure: Deininger, Jabbour, Mauro and Radich report no relevant financial disclosures.

 

Should discontinuation of TKI therapy be offered to patients with complete molecular response outside of a clinical trial?

Yes, discontinuation should be offered to select patients at low risk for relapse. However, selecting such patients might not be so simple.

The community oncologist who takes care of these patients should be able to offer discontinuation of TKI therapy in selected patients with CML. There are several reasons for this.

For one, TKIs are not completely benign. There are side effects that make taking this therapy difficult for some patients. One of my family members father has CML and was on imatinib for a decade. He had some side effects — such as pancytopenia, rash and gastrointestinal distress — although all of these have been manageable. He achieved complete molecular remission within 3 months, with undetectable BCR-ABL transcripts by quantitative PCR, and has had a remission for about 9 years. I think he would be a good candidate to stop TKI therapy.

Second, there are a number of prospective clinical trials that showed you can take patients off TKI therapy with a maintained remission rate between 40% and 50%. In all trials in which TKI therapy was stopped, the vast majority of patients remain sensitive to TKI therapy, and achieved remission again after restoration of TKI therapy.

There are a couple of things we should think about when taking the patient off TKI therapy. At least two of the trials in which a large patient fraction remained in remission after TKI removal required greater than 4-log reduction in BCR-ABL transcripts for more than 2 years. This should be a minimum requirement for discussion of discontinuation.

The second-generation TKIs produce faster and deeper molecular remissions than first-generation imatinib. If speed and depth of molecular remission are important for maintaining a remission after discontinuation, then patients initially treated with these agents might be good candidates for discontinuation.

Obviously, the risks and benefits of discontinuation would have to be discussed with individual patients. Such a discussion should include the side effects of continuing the TKIs, which we sometimes underestimate. The standard of care has been to treat for a 2-log reduction within 3 to 6 months, but this is not enough for consideration of discontinuation of therapy. You need a rapid, 4-log reduction at least, and for the molecular remission to be sustained for a long time before consideration of discontinuation.

There have been several prognostic scoring systems for CML, such as the Sokal or EUTOS systems. Patients with poor prognostic scores in these systems should not be considered for discontinuation of therapy.

It is important to follow patients who have discontinued therapy more closely once they are taken off therapy. They need structured follow-up with quantitative PCR for BCR-ABL at least every 3 months.

 

Robert A. Hromas, MD, FACP, is chairman of the department of medicine at the University of Florida and Shands Hospital. He can be reached at P.O. Box 100277, Gainesville, Fla. 32610. Disclosure: Hromas reports no relevant financial disclosures.

Discontinuation is a sound, evidenced-based approach right now, but longer-term follow-up will be key.

To stop or not to stop imatinib therapy for CML is the most compelling question in CML therapy today. Three phase 2 studies (STIM-1, TWISTER and STIM-2) with relatively short follow-up have shown that discontinuing imatinib in chronic-phase CML patients in complete molecular remission (CMR) by international standard (IS) results in approximately 50% treatment-free survival. Most relapses occurred in the first 6 months after discontinuation, and all relapsing patients responded to re-initiation of imatinib.

The studies differed in that STIM-2-enrolled patients were treated exclusively with imatinib, whereas some patients in the two other studies received interferon prior to imatinib.

Currently, the National Comprehensive Cancer Network and European LeukemiaNet recommend discontinuing imatinib only in the context of a clinical trial. What should practitioners caring for patients with CML outside of an academic center do?

Let us consider another scenario: A new drug combination for the treatment of a malignancy demonstrated remarkable results in three phase 2 clinical trials. Should the community physician consider using this combination? Most would answer yes to this question. However, many phase 3 studies based on preceding phase 2 studies resulted in negative results.

Let us now look at CML: Three phase 2 studies suggest that in the appropriate scenario — that is, for patients in CMR for at least 2 years by IS — one can stop imatinib. Should the community physician treating CML consider stopping imatinib? The answer is yes, and with these additional, very important considerations: the physician must obtain informed consent from the patient; must follow up with the patient monthly for the first 6 months, and at least every 2 months thereafter for the first year; and must stay up to date on new studies involving imatinib discontinuation for treatment of CML.

Although this is a sound and evidence-based approach at this point, longer-term follow-up may show detrimental effects of imatinib discontinuation.

Meir Wetzler, MD, FACP, is professor of medicine and chief of the leukemia section in the department of medicine at Roswell Park Cancer Institute. He can be reached at Roswell Park Cancer Institute, Elm and Carlton streets, Buffalo, N.Y. 14263.  Disclosure: Wetzler reports no relevant financial disclosures.