Advancements in CLL Treatment: Optimizing Outcomes with Novel Therapeutics and Strategies​ - Episode null

BTK Inhibitors vs Venetoclax: Considerations for Frontline Treatment Selection in CLL

In Partnership With:

Partner | Cancer Centers | <b>The University of Texas MD Anderson Cancer Center</b>

Nitin Jain, MD, discusses the use of BTK inhibitor– and venetoclax-based regimens for the frontline treatment of chronic lymphocytic leukemia.

The treatment landscape for chronic lymphocytic leukemia (CLL) has evolved with multiple frontline options including regimens featuring covalent BTK inhibitors or venetoclax (Venclexta); selecting an appropriate therapy for an individual patient should account for their specific disease characteristics and personal treatment goals, according to Nitin Jain, MD.

“There are pros and cons [to both of these approaches], and at the end of the day, there also needs to be a discussion with the patient to see what their viewpoint is if all things are equal,” Jain said in an interview with OncLive®.

In the interview, Jain highlighted the importance of considering patient-specific factors when selecting a frontline treatment regimen for those with CLL, expanded on the management of adverse effects (AEs) associated with these different agents, and detailed ongoing research in the frontline and relapsed/refractory CLL settings.

Jain is a professor of medicine in the Department of Leukemia at The University of Texas MD Anderson Cancer Center in Houston.

OncLive: Given the current treatment paradigm for CLL, what is your preferred frontline treatment approach?

Jain: The CLL treatment [paradigm has] evolved rapidly in the last few years. Currently, at least in the United States, there are 2 different approaches you could undertake. [One option is to] use a continuous BTK inhibitor approach, which will [include] drugs such as ibrutinib [Imbruvica], acalabrutinib [Calquence], or zanubrutinib [Brukinsa], which are supposed to be taken continuously, daily, and indefinitely. In the context of acalabrutinib, it can also be combined with obinutuzumab [Gazyva] for a duration of 6 months.

The other approach is a venetoclax-based therapy, which combines venetoclax with obinutuzumab for the first 6 months. [This] is a time-limited approach with a 1-year duration of therapy.

There are pros and cons these approaches. Continuous daily therapy with a BTK inhibitor is logistically a bit easier. Within the context of venetoclax plus obinutuzumab, you need to monitor for tumor lysis syndrome, and there are more frequent visits to the doctor’s office in the first couple of months. There's also a risk for neutropenia [with venetoclax-based approaches], which can occur in about half of the patients.

On the flip side, the 1-year [fixed-duration] therapy with venetoclax plus obinutuzumab does not have a BTK inhibitor, so you don't have the AEs [associated with these agents], such as atrial fibrillation, hypertension, bleeding risk, arthralgias, and [others].

There are also some genomic considerations [for selecting a frontline regimen]. Patients who have 17p deletions or TP53 mutations, the recommendation currently will be for continuous BTK inhibitor therapy. From a patient standpoint, if a patient has big lymph nodes or abnormal creatine, venetoclax-based therapy may be more difficult to administer. [These factors may] not be a complete contraindication, but there may be more risk involved.

If a patient has a risk for cardiovascular issues, [such as] atrial fibrillation, heart attack, or stent placement, or if they are receiving anticoagulation, that may sway me away from using a BTK inhibitor [in favor of] a ventoclax-based therapy.

There are many patients who like the ease of a daily tablet to be taken for a long time. [Then] there are other patients, especially those on the younger side, who want a 1-year [fixed-duration] therapy, and then they can be off therapy for a substantial number of years.

Are there any agents currently under investigation in the relapsed/refractory CLL space?

We are seeing encouraging data in the relapsed/refractory space for CLL. We have drugs such as pirtobrutinib [Jayprica] that was FDA approved [in December 2023] based on [data from] the phase 1/2 BRUIN study [NCT03740529]. [In the study, investigators evaluated] single-agent pirtobrutinib, which is a noncovalent BTK inhibitor, and what they saw was that the majority of patients responded. On average, in this relapsed/refractory patient population, responses lasted for 1.5 to 2 years.

We also had data presented with lisocabtagene maraleucel [liso-cel; Breyanzi], a CAR T-cell therapy that was approved [by the FDA] in March 2024 for [patients with relapsed/refractory] CLL [after at least 2 prior lines of therapy, including a BTK inhibitor and a BCL-2 inhibitor]. [Liso-cel] is the first CAR T-cell [therapy] approved for CLL.[The data also] suggested that the complete response rate was 20%; [the overall response rate was] 45%.

[We are also] seeing some encouraging data with BTK degraders, which are a new class of drugs coming along. There were a couple of presentations at the 2024 EHA Congress—one from Nurix [NX-5948] and one from BeiGene [BGB-16673]. [These] BTK degraders are looking promising with short follow-up right now. Again, the majority of the patients have achieved a partial remission, which [is to be expected] with BTK degraders.

How could data for acalabrutinib plus venetoclax affect the frontline treatment paradigm in CLL?

In the frontline setting, the field is evolving. We are about to see some randomized studies being reported. We are looking forward to the 2024 ASH Annual Meeting, where it is expected that [data from] the [phase 3] AMPLIFY study [NCT03836261] will be reported. [This was] a randomized study of chemotherapy vs acalabrutinib in combination with venetoclax [with or without] obinutuzumab.

We don't know the [full] data for [AMPLIFY] trial yet, but we know from [a news] release that the trial is positive in favor of the experimental arms over chemotherapy. We'll have to wait until ASH to know how exactly this panned out. [These data] have the potential to be practice changing because they may lead to approval of acalabrutinib plus venetoclax, which [would provide] 2 oral drugs together for patients with CLL in United States for the first time.

How has the FDA approval of pirtobrutinib affected the relapsed/refractory CLL treatment paradigm?

Pirtobrutinib is a noncovalent BTK inhibitor; although it is a BTK inhibitor, it is different from ibrutinib, acalabrutinib, and zanubrutinib, which are the approved [covalent] BTK inhibitors. [These covalent agents] work with the same mechanism of action, where they form a covalent bond with C481S, and their mechanisms of resistance, in large part, are because of a resistance mutation, BTK C481S. However, pirtobrutinib does not need to bind to this site.

Pirtobrutinib can inhibit the BTK without binding to C481S; therefore, it can work for patients who have a resistance mutation to these other [covalent] BTK inhibitors and [be an option] for patients who have [progressed] on other BTK inhibitors. That's how the BRUIN study was designed; the majority of patients had received a prior covalent BTK inhibitor. Quite a substantial number of patients had also received prior venetoclax. [During BRUIN, investigators] saw high rates ORR and [durable] remissions.

When the FDA approved pirtobrutinib, it was approved for patients who had progressed on prior covalent BTK inhibitors and venetoclax, which was appropriate, because that's where most of the data [from BRUIN was derived]. That has provided an important strategy, because this group of double-exposed or double-refractory patients did not have many good therapies [available]. We had PI3K inhibitors approved for that setting, but those agents have their own toxicities; CAR T-cell therapy was only recently approved for that group of patients; and we have other investigational clinical trials or allogeneic stem cell transplant. However, pirtobrutinib is a once-per-day pill, which can provide a high level of remission—mostly partial remissions—but good remissions for majority of the patients being treated with pirtobrutinib.

[This information] provided an important sequencing strategy [for patients with relapsed/refractory CLL] after progression on a covalent BTK inhibitor. Many of the trials with pirtobrutinib are now looking to introduce it in earlier lines of therapy, including the frontline setting, and those trials are ongoing.

How do you see the CLL treatment paradigm evolving in the coming years?

We have already moved away from chemotherapy, and now, [we could look at] what has been done in the context of multiple myeloma, where they're using their best drugs up front in [combination therapies]. That’s where the field of CLL [could be] moving, where we combine our best drugs up-front, whether it's a BTK inhibitor with a BCL-2 inhibitor or a BTK inhibitor plus a BCL-2 inhibitor and an anti-CD20 antibody. Those are the 3 classes of drugs we [currently] have.

As I mentioned, there are several ongoing phase 3 studies looking at these different combinations. Some of them we may hear at ASH in 2024; some [data may be read out] next year. In the future, maybe a few years down the line, we could be doing mostly [targeted combination therapy] for a time-limited duration, whether it's 1 or 2 years, or some kind of minimal residual disease [MRD]–guided therapy, as opposed to a continuous, lifelong BTK inhibitor, which is 1 of the current strategies.

What are some of the key management strategies for AEs associated with BTK inhibitor– or venetoclax-based therapies?

If a patient is receiving a venetoclax-based therapy, neutropenia can occur in approximately half of patients; in those patients, you will have to hold venetoclax and give granulocyte colony-stimulating factor support. When you restart venetoclax, sometimes it will be at the same dose, and sometimes we dose reduce to see if the neutropenia will recur. If you're using obinutuzumab, you have to be aware of infusion reactions, which can occur with the first dose and is very common.

Subsequent generations of BTK inhibitors have become safer; ibrutinib was a first-generation BTK inhibitor, but it has fallen out of favor [because of its associated] toxicities. However, BTK inhibitors can have toxicities such as atrial fibrillation, arthralgias, myalgias, skin rash, bleeding complications, and hypertension. Those are the things you have to be aware of.

Again, some of these [AEs] will require co-management with a cardiology team. For example, with atrial fibrillation, the cardiology team may want to add anticoagulation, and then the risk of bleeding increases if you're using a BTK inhibitor. Sometimes you have to individualize the therapy where there is a risk of bleeding. Let's suppose a patient also has low platelet count. Do you want to continue the BTK inhibitor, give a treatment free-interval, or move to a venetoclax-based therapy?

Those are individual decisions we have to make for patients. I don't think there could be set guidelines on what to do [for each individual]. However, important clinical complications do occur and require management from a cardiology team.

Within the context of CLL, how is treatment intolerance and resistance defined?

Intolerance is somewhat straightforward. If, for example, a patient is taking ibrutinib and having an AE—whether it's a hypertension, diarrhea, skin rash, or joint aches—you could consider decreasing the dose of the drug first before stopping the drug for intolerance. However, intolerance would be [defined] as stopping [treatment] due to AEs.

For treatment failure or resistance, if a patient is taking a BTK inhibitor and their [white blood cell counts] start rising, their lymph nodes start progressing, then it will be clear they are [progressing on] the BTK inhibitor.

However, sometimes a patient may be on a BTK inhibitor, have some AEs, stop the BTK inhibitor, and then a month or 2 later, the disease starts progressing. Now, that patient’s disease could get labeled as progression on a BTK inhibitor, but technically, they were not taking the BTK inhibitor when the disease relapsed. In those patients, you have to take a careful look at their history to see whether they were truly intolerant or truly resistant [to the treatment]. If a patient who had stopped for intolerance and then had their disease progress, I would say that patient is more intolerant than resistant.

With venetoclax, it can be a bit challenging to define failure, and I don't think there isclear guidance there since venetoclax-based therapies are typically given for 1 to 2 years. How do you define failure there [if a patient finished treatment]? Some people say if progression happens within 1 to 2 years of stopping the drug, that should be considered [treatment] failure. However, as a CLL community, I don't think we have standard definitions for treatment failure yet in the context of venetoclax.

What additional research or advancements are needed to enhance treatment strategies within the current CLL paradigm?

We have made remarkable progress in the context of CLL. The strategies today are much safer than what they were 5 to 10 years ago; however, we certainly need to fine tune treatment [further].

There are still lots of unanswered questions, such as what is the best BTK inhibitor to combine? Should it be combined with BTK degraders? Is venetoclax the best BCL-2 inhibitor because that's the only approved agent? If you were to combine these drugs, what should the end point be? How long should these drugs be continued? Should we be continuing them for 1 or 2 years, or based on MRD guidance?

Some of those [questions] are being addressed in [ongoing] randomized trials, whereas others are still out in the open. We don't fully know the answers to those questions. Some of these things [can be addressed by] fine-tuning treatment, including treatment based on individual genomics. For example, should a patient with mutated IGHV get the same duration of treatment as a patient with unmuted IGHV, who is a higher-risk patient? What about patients with 17p deletions? Should they be getting a different kind of treatment? A lot of effort in the field is trying to fine-tune and tailor treatment to specific genomic subgroup of patients with CLL.