B-cell receptor signaling is important to the pathogenesis and progression of B-cell malignancies, with BTK playing a crucial role in the signaling pathway. Therefore, targeting BTK has proven a useful treatment strategy for managing patients with CLL/SLL.

The class of BTK inhibitors was pioneered with the covalent BTK inhibitors including ibrutinib, acalabrutinib, and zanubrutinib. Covalent BTK inhibitors bind irreversibly to the catalytic ATP-binding domain of the BTK protein on cysteine 481 (C481), thereby blocking autophosphorylation and downstream signaling through PLCγ2.^1,2

The next development in this therapeutic drug class was the introduction of noncovalent BTK inhibitors, which reversibly bind to the ATP pocket regardless of C481 mutation status.¹ To date, pirtobrutinib is the only noncovalent BTK inhibitor recommended in guidelines and FDA-approved for CLL/SLL in clinical use in the United States.³ Other noncovalent BTK inhibitors are in development, including fenebrutinib, nemtabrutinib, and vecabrutinib.²

The selectivity of the covalent inhibitors for BTK and other on-target and off-target kinases varies across the available BTK inhibitors (acalabrutinib, ibrutinib, and zanubrutinib). We see that acalabrutinib exhibits the highest selectivity for the BTK protein, followed by zanubrutinib, with ibrutinib being the least selective among the 3 covalent BTK inhibitors.⁴

A key objective in the development of more selective compounds targeting BTK has been to decrease off-target binding activity and to improve the AE profile of this drug class.

Although the available covalent BTK inhibitors are very efficacious, and some have been shown to provide prolonged overall survival (OS) in clinical trials, as I discuss later in this module, patients eventually develop resistance to covalent BTK inhibitors and experience disease progression.^5-7

The most frequent cause of covalent BTK inhibitor resistance is gene mutations that affect the C481 site of the BTK protein, where the covalent BTK inhibitors bind.^6,8 Mutations at C481 abrogate the covalent binding, conferring resistance to covalent BTK inhibitors and restoring the ATP-binding activity of BTK. Because of their short half-life in plasma, covalent BTK inhibitors can only achieve incomplete inhibition of the BTK pathway in patients with a confirmed C481 mutation.^7,9 Of note, mutations in other locations on BTK occur much more rarely during treatment with these covalent BTK inhibitors.²

The mechanism of resistance to covalent BTK inhibitors via the emerging mutations at C481 led to the development of next-generation noncovalent BTK inhibitors that do not rely on C481, such as pirtobrutinib.^10-12

Pirtobrutinib is a selective, noncovalent (reversible) BTK inhibitor, currently approved by the FDA for treating adults with CLL or SLL who have received at least 2 previous lines of therapy, including a BTK inhibitor and a BCL-2 inhibitor.³

If we look at the kinome selectivity map for pirtobrutinib, we can see high selectivity for BTK. Moreover, pirtobrutinib also has a long plasma half-life, allowing for complete BTK target occupancy with once-daily dosing.^13,14

I would like to briefly review the efficacy of these BTK inhibitors starting with the phase III ALLIANCE A041202 study evaluating ibrutinib alone or ibrutinib plus rituximab (IR) vs bendamustine plus rituximab (BR) in the first-line treatment of older or physically unfit patients with CLL (N = 547).^15,16 BR was administered for 6 cycles whereas ibrutinib was administered as continuous therapy until disease progression occurred. The BR arm allowed crossover to the ibrutinib arm after disease progression. The primary endpoint of the study was PFS.

After a median follow-up of 55 months, data for median PFS showed that ibrutinib alone and IR were superior to BR (HR for ibrutinib vs BR: 0.36; P <.001; HR for IR vs BR: 0.36; P <.001).¹⁶ Median PFS was NR for the arms containing ibrutinib and was 44 months for the BR arm. The 48-month PFS rates were 76% for both ibrutinib-containing arms vs 47% for the BR arm.

All patient subgroups also showed improved PFS with IR vs BR, except patients with mutated IGHV disease.

Taken together, these data showed that the addition of rituximab to ibrutinib was not superior to ibrutinib alone, so it is not a standard treatment addition to ibrutinib.

BR treatment demonstrated the expected safety profile, consisting mostly of hematologic toxicity and infections.¹⁶ In the study arms containing ibrutinib arms, hematologic toxicity also occurred but at a lower frequency. With ibrutinib, we also see the class-associated AE of bleeding.^17,18 However, higher-grade bleeding events with ibrutinib occurred infrequently (grade 3/4: 1%/1%).

Rates of grade 3 atrial fibrillation (ibrutinib vs IR vs BR: 8% vs 6% vs 3%, respectively) and hypertension (ibrutinib vs IR vs BR: 29% vs 33% vs 14%, respectively) were twice as high in the ibrutinib-containing arms compared with the BR arm. These AEs are also associated with the BTK inhibitor class, so we need to closely monitor for them when treating patients with BTK inhibitors.

When discussing the safety profiles of the ALLIANCE A041202 trial, I must note that the capture period of AEs here is different for the BR and ibrutinib arms because BR is a time-limited treatment and corresponding AEs were captured during the 6 cycles of treatment.¹⁹ Only unsolicited AEs were recorded after the 6 cycles. By contrast, for the 2 ibrutinib arms, AEs were solicited and captured for the entire duration of treatment. These facts should be considered when looking at these AE data. 

The next development in the BTK inhibitor drug class was the next-generation agent acalabrutinib, a more selective covalent BTK inhibitor that became available based on the results of the randomized phase III ELEVATE-TN trial. This trial compared continuous treatment with acalabrutinib alone or continuous acalabrutinib plus obinutuzumab (AO) for 6 cycles vs OC for 6 cycles in older patients with CLL or physically unfit patients with CLL and decreased renal function or other comorbidities (N = 535).^20-22 The primary endpoint was PFS by independent review committee (IRC) assessment.

The primary endpoint was met, with acalabrutinib and AO demonstrating clinically meaningful and statistically significant improvement in median PFS vs OC (HR for acalabrutinib vs OC: 0.24; P <.0001; HR for AO vs OC: 0.14; P <.0001).²² In addition, median PFS was higher with acalabrutinib-containing regimens vs OC in all patient subgroups, including those with or without del(17p), TP53, or IGHV mutations.

Of note, there was an improvement in PFS with the addition of obinutuzumab to acalabrutinib vs acalabrutinib monotherapy (HR: 0.58, P = .0229). The median PFS was NR in both of the arms containing acalabrutinib and was 27.8 months in the OC arm.

When we look at the safety analysis, we should again consider the impact of different treatment durations on the capture of AE data. In ELEVATE-TN we still see the AEs associated with BTK inhibitors in the acalabrutinib-containing arms, such as cardiac events, bleeding, and hypertension, but at a lower frequency and severity (fewer grade ≥3). Of note, the addition of obinutuzumab to acalabrutinib increased the incidence of grade 3 neutropenia (30.9% with obinutuzumab vs 11.2% without obinutuzumab) and infection (23.6% vs 16.2%, respectively). 

Next, I would like to move on to discuss data from the multicohort phase III SEQUOIA study in patients with previously untreated CLL/SLL who meet the International Workshop on Chronic Lymphocytic Leukemia criteria for treatment (N = 479).^23,24 Here I focus on cohort 1, in which patients were randomized to receive continuous zanubrutinib or BR for 6 cycles. This trial accepted patients with CLL/SLL who were older, had comorbidities, or were unsuitable for fludarabine/cyclophosphamide/rituximab treatment (FCR). Cohort 1 only included patients without del(17p). The primary endpoint was PFS by IRC assessment.

At the 5-year update, patients receiving continuous zanubrutinib therapy had significantly better PFS by IRC assessment (the primary endpoint) compared with patients receiving BR (HR: 0.29; P <.0001), with an estimated 60-month PFS rate of 75.8% vs 40.1%, respectively.²³ The median PFS was NR in the zanubrutinib arm and was 44.1 months in the BR arm.

In addition, PFS was significantly better with zanubrutinib regardless of patients’ IGHV mutation status (mutated IGHV HR: 0.40; P = .0003; unmutated IGHV HR: 0.21; P <.0001).

The safety profile of zanubrutinib was encouraging. The rate of serious grade ≥3 AEs occurred at a lower frequency in the zanubrutinib arm compared with the BR arm (25% vs 39%, respectively) and regardless of the length of time that AEs were monitored.²³ However, any-grade diarrhea (16% vs 11%, respectively), hypertension (13% vs 8%, respectively), bleeding/hemorrhage (exposure-adjusted incidence rate per 100 person-months: 1.66 vs 0.35, respectively), and atrial fibrillation/flutter (exposure-adjusted incidence rate per 100 person-months: 0.13 vs 0.09, respectively) were more common with zanubrutinib vs BR, as expected.

To provide some context, I would also like to highlight data for other standard-of-care options in CLL/SLL, including data from the randomized phase III CLL14 trial evaluating time-limited therapy with the BCL-2 inhibitor venetoclax plus obinutuzumab vs OC in patients with previously untreated CLL and coexisting medical conditions (N = 432).^25,26 Obinutuzumab was administered for six 28-day cycles, chlorambucil for 12 cycles, and venetoclax was administered with step-up dosing during the first 2 of 12 total cycles. The primary endpoint of the study was PFS and a key secondary endpoint was OS.

After a median follow-up of 76.4 months, we see that the venetoclax plus obinutuzumab arm had a statistically significant prolonged median PFS vs the OC arm (PFS: 76.2 vs 36.4 months; HR: 0.4; P <.0001). By contrast, and although median OS was NR in both arms, no statistically significant difference in median OS was reported (HR: 0.69; P = .052).

Looking at PFS results for subgroups defined by TP53 status and IGHV status, here we also see that limited duration treatment with venetoclax plus obinutuzumab vs OC resulted in a median PFS benefit across subgroups, particularly for the higher-risk subgroups of TP53 mutation status (51.9 vs 20.8 months, respectively; HR: 0.56) and IGHV unmutated status (64.8 vs 26.9 months, respectively; HR: 0.30).²⁶

These data may also be useful to inform the prognostic impact of del(17p) and/or TP53 mutation on shorter PFS compared with the PFS in patients without del(17p) and/or TP53 mutation and in those with unmutated IGHV vs mutated IGHV for either of the treatment arms evaluated.

The most common AEs in both treatment arms were hematologic and included neutropenia, thrombocytopenia, and anemia.²⁵

I would also note here that tumor lysis syndrome (TLS), a serious potential AE with venetoclax, requires prophylaxis and step-up dosing of venetoclax when initiating therapy to reduce the risk of patients developing TLS as much as possible.

Cross-trial comparison of survival outcomes data is limited by many factors that contribute to variability in trial conduct. In clinical practice, however, we must do cross-trial comparisons because in the first-line setting, we do not have head-to-head clinical trial comparisons for the various treatment options. Looking at a chart of the approximate 4-year PFS rates of the different treatment regimens and trials to date in this disease setting, we see that there is no single best initial treatment in CLL.^{16,21,24,27-35}

Available data also suggest that PFS varies mostly in relation to the age and fitness of the patient(s) receiving treatment for CLL/SLL. For instance, 4-year PFS rates with percentages in the mid 70s are often seen in trials with a median patient age of approximately 70 years, whereas 4-year PFS rates with percentages in the mid 80s are associated with trials with a median patient age of 60 years (ie, in younger and more physically fit patients).

Ultimately, many of these first-line treatments achieve good outcomes, so multiple other factors influence our treatment choice including patient preference, comorbid conditions, toxicities, and efficacy data in patient subgroups.

Next, l discuss the relapsed/refractory (R/R) treatment setting, where we have the results of several hallmark phase III trials available. The RESONATE study of ibrutinib was the first to lead to market approval of a BTK inhibitor for R/R CLL.³⁶ Later, venetoclax plus rituximab was evaluated vs standard chemoimmunotherapy consisting of BR in the MURANO trial.³⁷ And there were head-to-head comparisons of zanubrutinib vs ibrutinib and acalabrutinib vs ibrutinib in the ambitious ALPINE and ELEVATE-RR trials, respectively.^38-40

The open-label phase III RESONATE trial evaluated ibrutinib vs ofatumumab in patients with R/R CLL/SLL who had received at least 1 previous treatment (N = 391).³⁶ In this trial, ibrutinib showed a dramatically superior improvement in median PFS vs ofatumumab (44.1 vs 8.1 months, respectively; HR: 0.148) and median OS (67.7 vs 65.1 months, respectively; HR: 0.810) and despite patient crossover to the ibrutinib arm from the control arm being allowed. Following the results from the RESONATE trial, at that time, ibrutinib replaced ofatumumab as the standard of care in R/R CLL/SLL. These results were significant also because of the patient treatment history of a median of 3 previous lines of therapy in the experimental arm and 2 in the control arm.

As previously mentioned, common AEs observed with BTK inhibitors include infections and neutropenia, but the frequency of these AEs decreased with time during the 6 years of follow-up.

We then saw results from a head-to-head trial of BTK inhibitors in the randomized, noninferiority phase III ELEVATE-RR trial comparing acalabrutinib vs ibrutinib, both as continuous therapy, in patients with high-risk R/R CLL, defined as having a del(17p) or del(11q) mutation (N = 533).^38,41 The primary endpoint of PFS showed no significant difference between the 2 agents and the noninferiority boundary for efficacy was met with an HR of 1.00 (95% CI: 0.79-1.27).

Regarding safety, however, acalabrutinib demonstrated a more favorable safety profile vs ibrutinib. The acalabrutinib arm had significantly fewer cardiovascular-related AEs, such as atrial fibrillation (9% vs 16%, respectively), hypertension (9% vs 23%, respectively), and other common AEs seen with BTK inhibitors such as diarrhea (35% vs 46%, respectively), arthralgia (16% vs 23%, respectively), and bleeding events (38% vs 51%, respectively). By contrast, acalabrutinib was associated with a higher incidence of headache (35% vs 20%, respectively) and cough (29% vs 21%, respectively) vs ibrutinib.⁴¹

The next key study was the randomized phase III ALPINE trial comparing zanubrutinib vs ibrutinib, both used as continuous therapy in patients with R/R CLL/SLL and at least 1 previous systemic treatment for CLL/SLL, excluding a previous BTK inhibitor (N = 652). The trial designs of ALPINE and ELEVATE-RR were similar but had some key differences.³⁹ First, the ALPINE study was not limited to high-risk patients. Second, the median number of previous lines of therapy in the ALPINE clinical trial (1) was lower than in ELEVATE-RR (2).^39,41 

The primary endpoint in ALPINE was met, with zanubrutinib demonstrating statistically significant and clinically meaningful PFS superiority compared with ibrutinib (HR: 0.65; 95% CI: 0.49-0.86; P = .0024) when assessed by both the investigator and by the IRC.

At an overall median follow-up of 42.5 months, the median PFS benefit with zanubrutinib vs ibrutinib was maintained (HR: 0.68; 95% CI: 0.54-0.84), including in patients with or without del(17p)/TP53 mutation. Moreover, the overall response rate was 85.6% vs 75.4% in favor of zanubrutinib.

Regarding safety, the AE profile of zanubrutinib appears to be more manageable compared with that of ibrutinib. Hypertension was observed with a similar frequency with zanubrutinib vs ibrutinib (27% vs 25%, respectively) but atrial fibrillation was substantially lower with zanubrutinib (7% vs 17%, respectively).

These 2 head-to-head studies of acalabrutinib (ELEVATE-RR) or zanubrutinib (ALPINE) vs ibrutinib suggest that acalabrutinib is overall better tolerated than ibrutinib across all class-associated AEs, with similar efficacy as ibrutinib, but with a higher incidence of headache.^39,41 By contrast, zanubrutinib appears to be more efficacious than ibrutinib, but with fewer advantages in the AE profile compared with ibrutinib. However, at this time, we do not have data directly comparing acalabrutinib and zanubrutinib to definitively determine differences in AE profiles.

Then we had the multicenter, open-label phase III MURANO trial evaluating venetoclax plus rituximab in R/R CLL/SLL, leading to its use in this setting. Patients in the experimental arm received 2 years of venetoclax treatment with six 28-day cycles of rituximab, whereas the control arm received 6 cycles of BR.^37,40,42

Median PFS at the 5-year follow-up was significantly prolonged in the arm that received venetoclax plus rituximab, at 54.7 months vs 17.0 months for patients receiving BR (HR: 0.19; P <.0001), and was similarly longer at the 7-year follow-up. With a median follow-up of 99.2 months, median OS had not been reached in the venetoclax plus rituximab arm and was 87.8 months for patients receiving BR (HR: 0.53; 95% CI: 0.37-0.74). 

The AE profile for venetoclax was as expected, with neutropenia being the most common AE in the experimental arm and neutropenia causing the most dose interruptions because of an AE.^42-44 The most common grade ≥3 AEs in the venetoclax plus rituximab arm and the BR arm included neutropenia (62% and 44%, respectively), anemia (11% and 14%, respectively), and pneumonia (7% and 10%, respectively).

Most of the trials that I have reviewed thus far in the R/R setting did not allow for prior BTK inhibitor or BCL-2 inhibitor treatment. Thus, there was an unmet need to develop treatments for patients who had previously been exposed to BTK or BCL-2 inhibitors.

The open-label, randomized phase III BRUIN CLL-321 trial evaluated pirtobrutinib, a next-generation, noncovalent BTK inhibitor, vs idelalisib plus rituximab, with both treatments as continuous therapies, or BR. This trial enrolled patients with CLL/SLL who had previously been treated with a covalent BTK inhibitor, including those with a previous history of atrial fibrillation (N = 238).^45,46 In this study, patients in the idelalisib/rituximab or BR arm had the option to cross over to the pirtobrutinib arm if disease progression was confirmed by the IRC. The primary endpoint of the study was PFS by IRC assessment. Approximately one half of patients in this trial had also previously received a BCL-2 inhibitor (eg, venetoclax-based treatment).

The BRUIN CLL-321 trial met its primary endpoint of showing longer median PFS by IRC assessment with pirtobrutinib treatment vs idelalisib plus rituximab or BR treatment (14.0 vs 8.7 months, respectively; HR: 0.54; 95% CI: 0.39-0.75; P = .0002).⁴⁵

Additional analyses from the BRUIN CLL-321 trial demonstrated the benefit of pirtobrutinib vs idelalisib plus rituximab or BR across clinical and biological subgroups.⁴⁵ Of note, median PFS was also longer for patients in the pirtobrutinib arm who had previously received BCL-2 inhibitor–based treatment.

Results from the BRUIN CLL-321 clinical trial led to accelerated FDA approval of pirtobrutinib for adult patients with CLL or SLL who have received at least 2 previous lines of therapy, including a BTK inhibitor and a BCL-2 inhibitor.³

The AEs associated with pirtobrutinib in the BRUIN CLL-321 trial patient population were generally manageable. Of note, pirtobrutinib was associated with more instances of any-grade infections compared with idelalisib plus rituximab or BR (63.8% vs 49.5%, respectively), but fewer instances of neutropenia (26.7% vs 33.9%, respectively) and diarrhea (16.4% vs 31.2%, respectively).⁴⁵

The most common AEs of interest in the pirtobrutinib arm of the study were infection (63.8%), neutropenia (26.7%), and bleeding (21.6%). The incidence of grades ≥3 bleeding events was low (3.4%). The higher rates of grade ≥3 neutropenia (20.7%) and infection (29.3%) were not unexpected and likely occurred because this study had a heavily previously treated patient population.