Acalabrutinib monotherapy for treatment of chronic lymphocytic leukaemia (ACE-CL-001): analysis of the Richter transformation cohort of an open-label, single-arm, phase 1–2 study

Summary

Background

Patients with chronic lymphocytic leukaemia who progress to Richter transformation (diffuse large B-cell lymphoma morphology) have few therapeutic options. We analysed data from the Richter transformation cohort of a larger, ongoing, phase 1–2, single-arm study evaluating the safety and activity of the selective, irreversible Bruton’s tyrosine kinase inhibitor acalabrutinib for the treatment of chronic lymphocytic leukaemia or small lymphocytic lymphoma.

Methods

For this open-label, single-arm, phase 1–2 study, patients aged 18 years or older with biopsy-proven treatment-naive or previously treated diffuse large B-cell lymphoma (Richter transformation) or prolymphocytic leukaemia transformation (Eastern Cooperative Oncology Group performance status ≤2) were assigned to receive oral acalabrutinib 200 mg twice daily as monotherapy until disease progression or toxicity. Patients were enrolled across seven centres from four countries. Safety and pharmacokinetics were assessed as primary endpoints; secondary endpoints were overall response rate, duration of response, and progression-free survival. Safety was assessed in the all-treated population (patients who received ≥1 dose), and activity was assessed in the all-treated population (for progression-free survival) and efficacy-evaluable population (for response rate; patients in the all-treated population with ≥1 response assessment after the first dose). This trial is registered with ClinicalTrials.gov (NCT02029443).

Findings

Between Sept 2, 2014, and April 25, 2016, 25 patients with Richter transformation were enrolled; 12 (48%) were male and 23 (92%) were White. As of data cutoff (March 1, 2021), two (8%) of 25 patients remained on acalabrutinib. The median time on study was 2·6 months (IQR 1·8–8·4). The most common adverse events (all grades) were diarrhoea (12 [48%] of 25 patients), headache (11 [44%]), and anaemia (eight [32%]). The most common grade 3–4 adverse events were neutropenia (seven [28%] of 25) and anaemia (five [20%]). The most common reason for treatment discontinuation was disease progression (17 [68%] of 25 patients). 11 (44%) deaths were reported within 30 days of the last acalabrutinib dose; none was considered treatment-related. Acalabrutinib was rapidly absorbed and eliminated, with similar day 1 and day 8 exposures. The overall response rate was 40·0% (95% CI 21·1–61·3), with two (8%) of 25 patients having a complete response and eight (32%) having a partial response; the median duration of response was 6·2 months (95% CI 0·3–14·8). Median progression-free survival in the overall cohort was 3·2 months (95% CI 1·8–4·0).

Interpretation

Acalabrutinib appears to be generally well tolerated, although progression-free survival was relatively poor in this cohort of patients with Richter transformation. On the basis of these findings, the use of acalabrutinib monotherapy in this setting is limited; however, further assessment of acalabrutinib as part of combination-based regimens for patients with Richter transformation is warranted.

Funding

Acerta Pharma, a member of the AstraZeneca Group.

Introduction

Chronic lymphocytic leukaemia is among the most common adult leukaemias in most North American and European countries.¹ Outcomes for patients with chronic lymphocytic leukaemia have improved substantially with the introduction of new targeted therapies directed at inhibiting Bruton’s tyrosine kinase (BTK; ibrutinib or acalabrutinib)^2,3 and downstream B-cell receptor signalling or BCL-2 (venetoclax)⁴ as monotherapy and in combination.⁵ However, even with these advances, approximately 2–10% of patients with chronic lymphocytic leukaemia undergo disease progression to diffuse large B-cell lymphoma during their disease course, also known as Richter transformation.⁶

Therapy for Richter transformation should generally be administered immediately and traditionally included intensive chemoimmunotherapy regimens similar to those used in de novo diffuse large B-cell lymphoma.⁷ These regimens have been associated with low overall response, substantial toxicities, and often short remissions for most patients with Richter transformation, except for patients with clonally unrelated disease.^8,9 The median overall survival for patients with Richter transformation is approximately 8 months,¹⁰ with data suggesting a potentially worse overall survival for patients progressing after BTK inhibitor or venetoclax therapy.^8,11 Collectively, these results signify an unmet need for more effective treatments for Richter transformation.

Investigational therapies targeting Richter transformation have included multiagent cytotoxic therapies including oxaliplatin,¹² and checkpoint inhibitor therapy directed at programmed cell death protein 1 (PD1).¹³ These regimens generally provided modest clinical benefits. Acalabrutinib is a selective, next-generation, potent oral BTK inhibitor approved by the US Food and Drug Administration for the treatment of chronic lymphocytic leukaemia or small lymphocytic lymphoma and relapsed or refractory mantle cell lymphoma.¹⁴ Twice-daily dosing of acalabrutinib generated near-continuous BTK inhibition with high selectivity in an initial phase 1–2 study in a cohort of patients with previously treated chronic lymphocytic leukaemia.³ Here, we report the results from the cohort of patients with Richter transformation in that study.

Methods

Study design and participants

This open-label, single-arm, multicentre, phase 1–2 study (ACE-CL-001) enrolled a pre-specified cohort of patients aged 18 years or older with measurable chronic lymphocytic leukaemia (defined as ≥1 lymph node measuring ≥2 cm in the longest diameter), Eastern Cooperative Oncology Group (ECOG) performance status 2 or higher, and biopsy-proven transformation of chronic lymphocytic leukaemia to diffuse large B-cell lymphoma (Richter transformation) or to B-cell prolymphocytic leukaemia at baseline. These patients were enrolled at three centres in the USA, two centres in the UK, and two centres in Europe, one in Sweden and one in Italy (appendix p 1). Biopsies were reviewed by institutional pathologists and were not centrally reviewed. Patients with treatment-naive Richter transformation or previously treated Richter transformation were eligible regardless of previous therapy for chronic lymphocytic leukaemia; previous BTK inhibitor therapy for chronic lymphocytic leukaemia was allowed. Additional eligibility criteria have been previously published;³ full inclusion and exclusion criteria are summarised in the appendix (pp 2–3). All patients provided written informed consent and all institutional review boards approved the study protocol. The study was done according to the Declaration of Helsinki and the International Conference on Harmonisation Good Clinical Practice guidelines.

Procedures

Acalabrutinib 200 mg was administered orally twice daily in 28-day cycles until disease progression or unacceptable toxicity. Patients with disease progression were removed from the study. Dosing delays or modifications were based on the clinical judgement of the investigator. Patients who underwent haematopoietic stem-cell transplantation (HSCT) discontinued acalabrutinib before transplantation and could resume acalabrutinib treatment after transplantation.

Plasma samples for pharmacokinetic analysis were taken during cycle 1 pre-dose and at 0·25, 0·50, 0·75, 1·00, 2·00, 4·00, and 6·00 h post-dose on days 1 and 8 of cycle 1, as well as pre-dose on day 2. On days 15, 22, and 28, plasma samples were taken pre-dose and 1 h post-dose. Additional details about pharmacokinetic assessments are provided in the appendix (p 3). Plasma samples for pharmaco kinetic analyses were analysed with a validated liquid chromatography–tandem mass spectrometry assay method with an analytical range of 1–1000 ng/mL. A non-compartmental pharmacokinetic approach was used to analyse individual plasma acalabrutinib concentration-time data with Phoenix WinNonlin (version 6.4).

Assessments of the baseline incidence of mutations or chromosomal abnormalities were done in patients for whom data were available. Molecular markers were tested on blood cells by a central laboratory within 10 days before the first dose of acalabrutinib and included immunoglobulin heavy-chain variable (IGHV) mutation status; stimulated karyotype (metaphase cytogenetics); a chronic lymphocytic leukaemia fluorescence in situ hybridisation (FISH) panel [11q13, del(11)(q22·3), del(17)(p13·1), 13q, 6q22–23, BCL6, MYC]; NOTCH1, MYD88, SF3B1, and TP53 mutation analyses; and serum β2-microglobulin concentrations. Patients with three or more distinct chromosomal abnormalities present in more than one metaphase were considered to have a complex karyotype¹⁵ and those with IGHV gene sequences having more than 98% homology with the germline gene sequences were considered to have unmutated IGHV status. Assessment of a clonal relationship between diffuse large B-cell lymphoma and chronic lymphocytic leukaemia was not done.

CT with contrast (unless contraindicated) or PET-CT (preferred) examination of the chest, abdomen, and pelvis and any other disease areas was required for response assessment and was done at screening, at the end of cycles 2, 4, 6, 9, 12, 15, 18, 21, and 24, every 6 cycles thereafter, and as warranted based on investigator discretion.¹⁶

Adverse events were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.03, and their frequency, severity, and attribution were assessed and recorded at all visits. Adverse events of clinical interest were identified on the basis of non-clinical findings, emerging data from other clinical studies related to acalabrutinib, and known pharmacological effects of other approved BTK inhibitors. These adverse events included cardiac events, cytopenias, bleeding or bruising, hepatotoxicity, hypertension, infections, interstitial lung disease or pneumonitis, second primary malignancies, and tumour lysis syndrome. Additional details about safety assessments are provided in the appendix (p 3).

Outcomes

The primary endpoints of the study were the safety and pharmacokinetics of acalabrutinib. Safety was assessed based on the frequency, severity, and attribution of adverse events (appendix p 88). The secondary endpoint of the study was tumour response, evaluated by the overall response rate, duration of response, and progression-free survival. The overall response rate was defined as the proportion of patients who had a complete response, complete response with incomplete marrow recovery, or partial response while on treatment before the initiation of new anticancer therapy or HSCT. Duration of response was defined as the time from the date of having the first complete response, complete response with incomplete marrow recovery, or partial response to the date of disease progression or death due to any cause, whichever occurred first. Response was assessed by investigators based on the 2014 Lugano Classification.¹⁶ Confirmation of a complete response required a PET–CT or CT followed by a bone marrow biopsy for patients with a radiological complete response.

The analysis of progression-free survival was based on investigator assessments and serial imaging. Progression-free survival was defined as the time from the date of the first dose to the date of first disease progression or death due to any cause, whichever occurred first. In the absence of disease progression or death, the patient was censored at the date of last adequate assessment (censoring date). If a patient received an autologous or allogeneic HSCT, the patient was censored at the date of transplantation. If a patient started a new anticancer therapy before disease progression or death, the patient was censored at the date of the last adequate assessment (defined as either physical examination and complete blood count or PET-CT and complete blood count) before receiving the new anticancer therapy. In the absence of an adequate assessment after the first dose of acalabrutinib, including patients without an adequate assessment before starting a new anticancer therapy, patients were censored at day 1.

Due to the aggressive nature of Richter transformation, the pharmacodynamics of drugs might differ in patients with Richter transformation compared with patients with chronic lymphocytic leukaemia. Measurement of the pharmacodynamic parameters of acalabrutinib was an exploratory endpoint of the study (methods are provided in the appendix pp 3–4). Assessment of biological markers of B-cell function was an additional exploratory endpoint of the study.

Statistical analysis

Safety and progression-free survival were assessed in the all-treated population, defined as all patients who received at least one dose of the study drug, and response was assessed in the efficacy-evaluable population, defined as patients in the all-treated population with at least one response assessment after the first dose. As per the protocol, the study tested the null hypothesis (overall response rate ≤10%) against the alternative hypothesis (overall response rate ≥35%); by use of Simon’s optimal 2-stage design, a total sample size of 30 patients per cohort was estimated to have 90% power to achieve a one-sided significance level of 0·025 or less. Descriptive statistics were used to summarise continuous variables and time to onset of adverse events and time to initial response. Numbers and percentages of patients were used to describe categorical variables. Time-to-event endpoints were estimated with the Kaplan-Meier method. Subgroup analyses based on baseline and disease characteristics, including previous therapy, were done post hoc in patients who had an overall response. All data were analysed with SAS software (version 9.4).

This trial is registered with ClinicalTrials.gov (NCT02029443).

Role of the funding source

Acerta Pharma, a member of the AstraZeneca Group, provided the study drug, contributed to the design of the study, data collection, data analysis, and data interpretation, and funded medical writing assistance.

Results

Between Sept 2, 2014, and April 25, 2016, 25 patients with Richter transformation were enrolled and included in the present analysis (figure 1). Data are presented as of March 1, 2021 (data cutoff). The present analysis excludes four patients (three with transformation of chronic lymphocytic leukaemia to B-cell prolymphocytic leukaemia and one with transformation of follicular lymphoma to diffuse large B-cell lymphoma) from a previous analysis (data cutoff date June 1, 2016)¹⁷ because these histologies are distinct from Richter transformation. Major deviations from the protocol occurred in nine patients: four patients received prohibited concomitant medication (three patients had an allogeneic HSCT and were censored for the progression-free survival analyses; one patient received high-dose corticosteroid), one patient had serious adverse events reported outside of the prescribed time window for reporting, three patients did not meet one of the eligibility criteria (one had no ECOG score reported, one had a long QT interval on fluoxetine and fluconazole [fluoxetine was not an allowed concomitant medication at the time of this patient’s participation in the study], one had a bodyweight ≤45 kg), and one patient had a deviation reported for the signed informed consent form (the document had tracked changes for the site number and investigator name). The median age was 66 years (IQR 58–73); common baseline genomic abnormalities were TP53 mutations (nine [39%] of 23), del(17p) (seven [28%] of 25), and TP53 aberrancy (TP53 mutation or del(17p), or both; ten [42%] of 24; table 1). Patients received a median of one previous systemic anticancer therapy (IQR 0–2) after diagnosis of Richter transformation and before initiating acalabrutinib treatment (appendix p 6). 11 patients received no previous therapy for Richter transformation and 14 were previously treated for Richter transformation. Among the 14 patients who previously received treatment for Richter transformation, the most common regimens were cyclophosphamide, doxorubicin (hydroxydaunorubicin), vincristine, and prednisone (CHOP) plus rituximab (five [36%] of 14) and ibrutinib (four [29%] of 14; appendix p 7). 11 (44%) of 25 patients previously received chemotherapy after diagnosis of Richter transformation (seven received a CHOP-based regimen; three received an etoposide phosphate, prednisone, vincristine, cyclophosphamide, and doxorubicin hydrochloride [EPOCH]-based regimen; and three received a platinum-based regimen in addition to or instead of CHOP-based or EPOCH-based therapy). Nine (36%) of 25 patients received ibrutinib alone or a regimen containing ibrutinib before diagnosis of Richter transformation, as did five (20%) of 25 patients after diagnosis of Richter transformation (before acalabrutinib treatment). Seven (28%) of 25 patients were receiving ibrutinib for treatment of chronic lymphocytic leukaemia at the time of Richter transformation diagnosis. The median time on study (from treatment start date to safety follow-up or end of treatment date) was 2·6 months (IQR 1·8–8·4). At the time of data cutoff, two (8%) of 25 patients remained on acalabrutinib treatment. One of these patients had a treatment duration of 63·0 months. The other patient had a treatment duration of 61·1 months; however, the patient received HSCT during the study and stopped acalabrutinib for approximately 5 months. Reasons for treatment discontinuation were progressive disease (17 [68%] of 25 patients), death (two [8%] of 25), physician decision (two [8%] of 25), and withdrawal by patient (two [8%] of 25). Three (12%) of 25 patients had dose reductions. 11 (44%) deaths were reported within 30 days of the last dose; none was considered treatment-related.

Figure 1: Trial profile.

Table 1:

Baseline characteristics of patients in the Richter transformation cohort

	Overall cohort (n=25)
Median age, years	66 (58–73)
Age ≥65 years	13 (52%)
Sex
Female	13 (52%)
Male	12 (48%)
Race
White	23 (92%)
Black or African American	1 (4%)
Other	1 (4%)
Median time from initial chronic lymphocytic leukaemia diagnosis to transformation, years	4·5 (2·7–9·6)
Median time from transformation to first dose of acalabrutinib, years	0·3 (0·1–0·8)
ECOG performance status
0	5 (20%)
1	14 (56%)
2	6 (24%)
Patients with cytopenia	19 (76%)
Absolute neutrophil count ≤1·5 × 10⁹ cells per L	4 (16%)
Haemoglobin ≤11 g/dL	14 (56%)
Platelets ≤100 × 10⁹ cells per L	13 (52%)
Platelets ≤50 × 10⁹cells per L	7 (28%)
Presence of B symptoms	3 (12%)
β2-microglobulin >3 mg/L	10/10 (100%)
Bulky disease  ≥5 cm	13/24 (54%)
≥10 cm	2/24 (8%)
Genomic status
Del(11q)	4/25 (16%)
Del(17p)	7/25 (28%)
Complex karyotype*	10/23 (43%)
Unmutated IGHV	14/18 (78%)
TP53 mutation positive	9/23 (39%)
TP53 aberrancy positive†	10/24 (42%)
NOTCH1 mutation positive	4/23 (17%)
SF3B1 mutation positive	4/23 (17%)
MYC aberrations	6/24 (25%)

Data are n (%), n/N (%), or median (IQR). ECOG=Eastern Cooperative Oncology Group. IGHV=immunoglobulin heavy-chain variable.

^*Patients with three or more abnormalities were considered to have a complex karyotype.

†Patients with del(17p) or TP53 mutations, or both, were considered to be TP53 aberrant.

For the primary endpoint of safety, of the 25 patients who received at least one dose of acalabrutinib, 24 (96%) had one or more adverse events (table 2). No adverse events resulted in treatment discontinuation. The most common adverse events of any grade were diarrhoea (12 [48%] of 25), headache (11 [44%] of 25), and anaemia (eight [32%] of 25; table 2). All cases of diarrhoea were grade 1–2. Nine (75%) of 12 patients had only one event of diarrhoea. Headaches were grade 1 in 6 (55%) of 11 patients, with no grade 3 or worse headaches reported. The median time to onset of a headache was 2 days (IQR 2–8). The median duration of headaches was 14 days (IQR 6–35).

Table 2:

Grade 1–2 treatment-emergent adverse events occurring in at least 10% of patients and all grade 3 or worse events in the overall cohort (n=25)

	Grade 1–2	Grade 3	Grade 4	Grade 5
Diarrhoea	12 (48%)	0	0	0
Headache	11 (44%)	0	0	0
Anaemia	3 (12%)	4 (16%)	1 (4%)	0
Fatigue	5 (20%)	2 (8%)	0	0
Neutropenia	0	3 (12%)	4 (16%)	0
Arthralgia	5 (20%)	1 (4%)	0	0
Contusion	5 (20%)	0	0	0
Cough	5 (20%)	0	0	0
Thrombocytopenia*	2 (8%)	1 (4%)	2 (8%)	0
Pleural effusion	4 (16%)	1 (4%)	0	0
Pyrexia	4 (16%)	1 (4%)	0	0
Back pain	2 (8%)	2 (8%)	0	0
Decreased appetite	3 (12%)	1 (4%)	0	0
Fall	3 (12%)	1 (4%)	0	0
Hypokalaemia	2 (8%)	2 (8%)	0	0
Nausea	3 (12%)	1 (4%)	0	0
Pain in extremity	3 (12%)	1 (4%)	0	0
Weight decreased	4 (16%)	0	0	0
ALT increased	2 (8%)	1 (4%)	0	0
Clostridium difficile colitis	1 (4%)	2 (8%)	0	0
Confusional state	3 (12%)	0	0	0
Constipation	2 (8%)	1 (4%)	0	0
Dehydration	1 (4%)	2 (8%)	0	0
Dizziness	3 (12%)	0	0	0
Dyspnoea	3 (12%)	0	0	0
Febrile neutropenia	0	3 (12%)	0	0
Hypercalcaemia	0	2 (8%)	1 (4%)	0
Insomnia	3 (12%)	0	0	0
Myalgia	3 (12%)	0	0	0
Neutrophil count decreased	0	1 (4%)	2 (8%)	0
Pneumonia	1 (4%)	1 (4%)	0	1 (4%)
Rash	2 (8%)	1 (4%)	0	0
Tachycardia	3 (12%)	0	0	0
Vomiting	3 (12%)	0	0	0
Acute kidney injury	0	0	2 (8%)	0
Asthenia	0	2 (8%)	0	0
Bronchitis	1 (4%)	0	1 (4%)	0
Hypoxia	0	2 (8%)	0	0
Infection	1 (4%)	1 (4%)	0	0
Respiratory tract infection	1 (4%)	1 (4%)	0	0
Sepsis	1 (4%)	0	0	1 (4%)
Blood sodium decreased	0	0	1 (4%)	0
Brain abscess	0	0	0	1 (4%)
Bronchopulmonary aspergillosis	0	1 (4%)	0	0
Clostridium difficile infection	0	1 (4%)	0	0
Cytomegalovirus test positive	0	1 (4%)	0	0
GVHD in gastrointestinal tract	0	1 (4%)	0	0
Inguinal hernia	0	1 (4%)	0	0
Kidney infection	0	1 (4%)	0	0
Malignant hypertension	0	0	1 (4%)	0
Mycobacterium avium complex infection	0	1 (4%)	0	0
Pain	0	1 (4%)	0	0
Respiratory failure	0	0	1 (4%)	0

Data are n (%). ALT=alanine aminotransferase. GVHD=graft-versus-host disease.

^*Preferred term: platelet count decreased.

Grade 3 or worse adverse events occurred in 21 (84%) of 25 patients; the most common grade 3 or worse adverse events (occurring in ≥15% of patients) were neutropenia (seven [28%] of 25) and anaemia (five [20%] of 25; table 2). Grade 3 or worse thrombocytopenia occurred in three (12%) of 25 patients. Three deaths due to adverse events were reported: a brain abscess (possibly fungal; onset 112 days after the first acalabrutinib dose and 9 days after the last dose), sepsis (onset 29 days after the first acalabrutinib dose and 9 days after the last dose), and lung infection (onset 112 days after the first acalabrutinib dose and 26 days after the last dose).

17 (68%) of 25 patients had serious adverse events. The most frequent serious adverse events (occurring in ≥5% of patients) were febrile neutropenia (three [12%] of 25 patients), hypercalcaemia (three [12%]), acute kidney injury (two [8%]), fatigue (two [8%]), infection (two [8%]), and sepsis (two [8%]).

Among the clinical events of interest, atrial fibrillation was reported in two (8%) of 25 patients; both events resolved and did not lead to treatment discontinuation. No grade 3 or worse cardiac adverse events were observed. Haemorrhagic events occurred in 11 (44%) of 25 patients; all were grade 1–2. The most commonly reported haemorrhagic events (occurring in ≥5% of patients) were contusion (five [20%] of 25), epistaxis (two [8%] of 25), injection-site bruising (two [8%] of 25), and petechiae (two [8%] of 25). Infections of any grade occurred in 15 (60%) of 25 patients; grade 3 or worse infections occurred in eight (32%) of 25 patients.

Exposure to acalabrutinib varied due to overall pharmacokinetic variability, and there was no major difference in pharmacokinetic parameters between day 1 and day 8 (appendix pp 8–9). Geometric mean, arithmetic mean, and median non-compartmental pharmacokinetic parameters were calculated for all eight patients with available data (appendix p 8). Acalabrutinib was rapidly absorbed, with median plasma T_max values of 0·99 h on day 1 and 0·85 h on day 8. Geometric mean C_max values were 1087 ng/mL (100%) on day 1 and 1561 ng/mL (52·3%) on day 8. Geometric mean acalabrutinib AUC_last values were 1677 ng/mL (84·2%) on day 1 and 1746 ng/mL (37·6%) on day 8. Exposure to acalabrutinib was similar on days 1 and 8, and pre-dose concentrations on days 8, 15, and 22 were generally less than 30 ng/mL, indicating low potential for accumulation. Acalabrutinib plasma concentrations measured at 1 h post-dose on days 1, 8, 15, and 22 were variable and relatively constant over time, indicating neither accumulation nor induction of acalabrutinib clearance. In the six patients with pharmacokinetic samples and best responses available, greater increases in C_max and AUC_last from day 1 to day 8 were observed in patients who had best responses of progressive disease than in those who had best responses of stable disease (appendix p 10). Acalabrutinib was relatively rapidly eliminated, with mean t_1/2 values of 0·80 h (12·2%) on day 1 and 0·76 h (17·4%) on day 8. Acalabrutinib mean apparent clearance/fraction absorbed (CL/F) values were 134 L/h on day 1 and 110 L/h on day 8 and acalabrutinib mean apparent volume of distribution/fraction absorbed (Vz/F) values were 162 L on day 1 and 123 L on day 8.

The overall response rate was 40·0% (95% CI 21·1–61·3); two (8%) of 25 patients had a complete response, and eight (32%) of 25 had a partial response (table 3). The median duration of response for the ten responding patients was 6·2 months (95% CI 0·3–14·8). Seven of the ten responding patients remained on therapy for more than 6 months (range 7·4–63·0), two of whom were still receiving acalabrutinib at the data cutoff (figure 2A). Three of the seven responders (two with a complete response and one with a partial response) received an allogeneic HSCT (responses were reached before transplantation) and then restarted acalabrutinib treatment, two of whom remained on acalabrutinib for another 7 months and 6 months. The third patient was still receiving acalabrutinib at the data cutoff after 61·1 months, approximately 48·5 months of which occurred after re-initiation of acalabrutinib after HSCT. The overall response rate was similar in the 14 patients who previously received treatment for Richter transformation (six [43%; 95% CI 17·7–71·1] of 14) compared with the 11 patients who had not previously received treatment for Richter transformation (four [36%; 10·9–69·2] of 11). The overall response rate was 25·0% (three of 12 patients; 95% CI 5·5–57·2) in the patients who were previously exposed to ibrutinib (median duration of response 1·8 months [95% CI 1·5–7·2]) compared with 53·8% (seven of 13 patients; 25·1–80·8) in the patients without previous ibrutinib exposure (median duration of response 9·2 months [0·3–not reached]). Among the seven patients who were receiving ibrutinib for treatment of chronic lymphocytic leukaemia at the time of Richter transformation diagnosis, the overall response rate was 14% (95% CI 0·3–57·9) with one of seven patients having a partial response. Among the five patients who did not previously receive any chronic lymphocytic leukaemia therapies before diagnosis of Richter transformation, one had a partial response, three had progressive disease, and response data were not available for one patient. None of the four patients with del(11q) had a response; these patients had all been treated previously with ibrutinib and were treated with acalabrutinib for 1, 3, 28, and 52 days.

Table 3:

Response to acalabrutinib in patients with Richter transformation

	Overall cohort (n=25)
Overall response rate (complete response plus partial response)	10 (40%; 95% CI 21·1–61·3)
Best response
Complete response	2 (8%)
Partial response	8 (32%)
Stable disease	3 (12%)
Progressive disease	10 (40%)
Unknown	2 (8%)
Median time to initial response, months (IQR)	1·9 (1·6–2·1)
Median duration of response, months (95% CI)	6·2 (0·3–14·8)

Figure 2: Time on treatment and duration of response and progression-free survival.

(A) Characteristics of response and progression as assessed by the investigator. Ongoing responses are denoted. Green stars indicate time on treatment when haematopoietic stem-cell transplantation was received. The median time on treatment was 2·6 months (IQR 1·8–8·4). Denotations of complete response (CR), partial response (PR), stable disease (SD), progressive disease (PD), or unknown (U) on each patient’s timeline indicate the responses at assessment timepoints. (B) Progression-free survival estimated with the Kaplan-Meier method in all treated patients.

The median progression-free survival for all patients was 3·2 months (95% CI 1·8–4·0; 21 of 25 events; figure 2B). Among the 14 patients previously treated for Richter transformation before receiving acalabrutinib, the median progression-free survival was 2·0 months (95% CI 1·7–11·1; 11 events); 11 patients who had not been previously treated for Richter transformation before receiving acalabrutinib had a median progression-free survival of 3·2 months (1·6–3·7; ten events; appendix pp 12–13). Among 20 patients who previously received treatment for chronic lymphocytic leukaemia before receiving acalabrutinib, median progression-free survival was 3·5 months (95% CI 1·8–4·0; 16 events) compared with 1·8 months (1·2–6·8; five events) in five patients who did not (appendix p 14). Median progression-free survival was 2·0 months (95% CI 1·2–3·7; ten events) in 12 patients previously exposed to ibrutinib and 3·7 months (1·8–11·1; 11 events) in 13 ibrutinib-naive patients (appendix p 15).

Pharmacodynamic and immunological findings are reported in the appendix (pp 4–5, 16–22). Briefly, BTK occupancy was near complete (99·4–99·6%) 4 h after dosing and ranged from 95·1% to 96·9% at drug trough. No clear trends in T-cell, B-cell, natural killer cell, and monocyte cell count changes were noted over time.

Discussion

Here, we report the results of, to our knowledge, the first cohort of patients with Richter transformation treated with acalabrutinib monotherapy in a phase 1–2 study. Data from this multicentre study show acceptable tolerability of acalabrutinib and an overall response rate of 40·0% (95% CI 21·1–61·3). However, a durable clinical response or stable disease lasting 6 months or more was observed in only six patients. Additionally, two patients remained on acalabrutinib as of the data cutoff date, with more than 60 months of treatment, one of whom received HSCT, suggesting that few patients might benefit from monotherapy. Considering more promising responses were observed with novel combination therapies with other agents,^18,19 further evaluation of acalabrutinib in combination with other therapies that might have clinical activity in this disease setting is warranted. Treatment with R-CHOP is being directly compared with R-CHOP combined with acalabrutinib 100 mg twice daily in the randomised phase 2 STELLAR trial in patients with newly diagnosed Richter transformation in the UK (NCT03899337).²⁰

High-grade haematological adverse events (neutropenia, anaemia, and thrombocytopenia) occurred in much fewer patients in the present study compared with some studies evaluating traditional chemotherapy-based regimens.^21,22 For example, a lower incidence of grade 3 or 4 cytopenia was observed in the present study than in a previous study evaluating oxaliplatin, fludarabine, cytarabine, and rituximab (OFAR) therapy in patients with Richter transformation (neutropenia in seven [28%] of 25 patients vs 25 [83%] of 30, anaemia in five [20%] of 25 vs 15 [50%] of 30, and thrombocytopenia in three [12%] of 25 vs 23 [77%] of 30)²¹ and the incidence of grade 3 or 4 cytopenia was also much lower in the present study than in a combined population of patients with Richter transformation (n=15), advanced chronic lymphocytic leukaemia with autoimmune cytopenia (n=19), or high-risk chronic lymphocytic leukaemia (n=26) receiving treatment with R-CHOP (neutropenia in 55% of patients, anaemia in 75%, and thrombocytopenia in 65%).²²

Notably, this study used twice the dose of acalabrutinib (200 mg twice daily) approved for the treatment of chronic lymphocytic leukaemia and relapsed or refractory mantle cell lymphoma (100 mg twice daily).¹⁴ The dose for the Richter transformation cohort described here was based on the safety of the acalabrutinib 200 mg twice-daily and 400 mg once-daily doses in the phase 1 dose-escalation portion of the present study in patients with relapsed or refractory chronic lymphocytic leukaemia.³ The twice-daily regimen, which is now used across the acalabrutinib clinical programme, is supported by preliminary results showing durable responses and near-complete BTK occupancy in a study of acalabrutinib in spontaneous canine lymphoma, an aggressive tumour similar to human diffuse large B-cell lymphoma.²³ Notably, despite the higher acalabrutinib dose and aggressive nature of Richter transformation, the observed safety profile was similar to that reported in a pooled analysis of patients treated with acalabrutinib across multiple haematological malignancies.²⁴ Serious adverse events were more common in the present analysis, which is probably because of the nature of the patient population studied and because symptoms of progression to Richter transformation can be difficult to distinguish from drug toxicity in a heavily pre-treated population.

Overall, the pharmacokinetic parameters of acalabrutinib observed in this study were variable in the first 6 h after dosing and indicated relatively rapid absorption and elimination and low potential for accumulation, consistent with the pharmacokinetic profile of acalabrutinib observed in patients with chronic lymphocytic leukaemia (ie, without Richter transformation).³ In general, the mean exposure observed in this cohort after treatment with acalabrutinib 200 mg twice daily is approximately two times higher than the mean exposure observed with the acalabrutinib dosing approved for the treatment of patients with chronic lymphocytic leukaemia (100 mg twice daily).¹⁴ Exposure increases on both day 1 and day 8 were similar to the increases observed with acalabrutinib 100 mg twice daily dosing.³ Based on the small number of patients with pharmacokinetic samples and recorded clinical responses in this study, further analyses evaluating the relationship between acalabrutinib exposure and clinical response are needed and will provide more information about appropriate dosing.

Acalabrutinib monotherapy response rates in the present study appear to be in line with those reported with combination therapies in patients with Richter transformation in the first-line setting.²⁵ The majority of previously treated patients in the present study had received chemotherapy in combination with a CD20 antibody before initiating this study. Notably, patients who had not previously received ibrutinib had substantially better efficacy than patients who had previous ibrutinib exposure (overall response rate 53·8% [95% CI 25·1–80·8] vs 25·0% [5·5–57·2]; median progression-free survival: 3·7 months [95% CI 1·8–11·1] vs 2·0 months [1·2–3·7]). Previous evidence suggested a non-deleterious effect of del(11q) on progression-free survival and overall response rate outcomes in patients with relapsed or refractory chronic lymphocytic leukaemia treated with ibrutinib.²⁶ In our study of heavily pre-treated patients with Richter transformation, none of the four patients with del(11q) had a response, although this finding should be interpreted with caution given the small number of patients with del(11q) and the short duration of treatment (range 1–52 days). Additionally, the presence of del(11q) might be less of a prognostic factor in the case of heavily pre-treated patients,²⁷ such as patients with Richter transformation, who have poorer outcomes in general. Cross-trial comparisons in patients with Richter transformation are difficult because of the heterogeneity of the patient population; nevertheless, our results confirm poorer outcomes in these patients compared with patients with chronic lymphocytic leukaemia and support an urgent ongoing medical need for more effective regimens in this setting. Three patients who received acalabrutinib in this study were successfully bridged to allogeneic HSCT, which is an important therapeutic option in selected, fit patients with this disease and represents a potentially curative approach. Other approaches, such as chimeric antigen receptor T-cell therapy, might also be considered in this setting based on preliminary reports,²⁸ although a larger clinical trial will be required to ascertain the benefit of this therapeutic approach in patients with Richter transformation.

Several novel agents and therapy combinations are currently under investigation for Richter transformation.¹⁹ Data are emerging for newer classes of agents such as PD1 monoclonal antibodies in patients with Richter transformation; however, sample sizes are typically considerably smaller than in this study. Preliminary data from the phase 2 study of the PD1 monoclonal antibody pembrolizumab showed activity in nine heavily pre-treated patients with Richter transformation (overall response rate 44% [95% CI 14–79]; four of nine patients in the Richter transformation cohort), similar to the overall response rate for acalabrutinib in this study, but also showed higher rates of drug-related grade 3–4 thrombocytopenia, anaemia, and neutropenia (each reported in five [20%] of 25 patients in the combined relapsed or refractory chronic lymphocytic leukaemia [n=16] and Richter transformation [n=9] cohorts) with pembrolizumab compared with the rates observed in the present study with acalabrutinib.¹³ Comparatively, a recent study evaluating the BCL-2 inhibitor venetoclax combined with R-EPOCH reported an overall response rate of 76%,¹⁸ suggesting that novel combination strategies might be beneficial in this disease setting and warrant further investigation. Based on the results of our present study, the use of acalabrutinib monotherapy in Richter transformation is questionable given other available (although limited) options. Results of the phase 2 STELLAR study evaluating acalabrutinib and R-CHOP in patients with newly diagnosed Richter transformation²⁰ might shed more light on the potential benefit of acalabrutinib as part of combination therapy in patients with Richter transformation.

Limitations of our study include lack of assessment of the clonal relationship between chronic lymphocytic leukaemia and diffuse large B-cell lymphoma (Richter transformation) and the use of institutional pathologists instead of central pathology review. This limitation might be relevant given the documented diagnostic challenges involved in diagnosing Richter transformation.²⁹ We were also unable to test for BTK and PLCG1 mutations at the time of development of Richter transformation. Additionally, the use of univariate analysis in assessment of progression-free survival should be interpreted with caution due to the small number of patients in this study.

In summary, the results of this phase 2 study support the need for further investigation of acalabrutinib as part of combination-based regimens for Richter transformation and in patients with Richter transformation who are not candidates for combination therapy due to their advanced age and the presence of comorbidities.

Research in context.

Evidence before this study

We searched PubMed on August 24, 2020, using the search terms “treatment” occurring in the title or PubMed abstract AND “Richter” occurring in the title, with no restrictions on language or publication date. We found a paucity of information available about the activity and safety of targeted treatments in patients with Richter transformation. A minority of studies evaluated targeted therapies, most commonly inhibitors of Bruton’s tyrosine kinase, programmed cell death protein 1 (PD1), or CD20, but generally included small numbers of patients with Richter transformation or were isolated case reports. Moreover, the effectiveness of targeted treatments as well as the duration of activity remain unclear based on these reports, representing a considerable unmet need within this population.

Added value of this study

This phase 1–2 study is, to the best of our knowledge, the first to prospectively evaluate the activity and safety of a targeted treatment (the highly selective Bruton’s tyrosine kinase inhibitor acalabrutinib) in a specific cohort of patients who underwent Richter transformation. Moreover, given that only approximately 2–10% of patients with chronic lymphocytic leukaemia undergo disease progression to diffuse large B-cell lymphoma, this study included a relatively large number of patients with Richter transformation (n=25). In this study, acalabrutinib monotherapy was well tolerated but yielded an overall response rate of 40·0% (95% CI 21·1–61·3) and a relatively short median progression-free survival of 3·2 months (1·8–4·0) in patients with this aggressive disease.

Implications of all the available evidence

These results suggest that acalabrutinib has acceptable tolerability with documented responses in patients with Richter transformation; however, progression-free survival was relatively poor. Considering the improved response with novel combinations of other agents, this study provides justification for evaluating acalabrutinib in combination with other therapies that might have clinical activity in this disease setting.

Data sharing

Data underlying the findings described in this Article can be obtained in accordance with AstraZeneca’s data sharing policy.