Zanubrutinib: Discovery, Development, Safety and Pharmacovigilance
Zanubrutinib is an oral, covalent inhibitor of Bruton tyrosine kinase (BTK), developed for B-cell malignancies. Its development illustrates how a medicinal chemistry programme can improve target selectivity without eliminating mechanism-related hazards. The scientific story extends from the role of BTK in B-cell receptor signalling, through tissue-occupancy studies and indication-specific clinical trials, to continuous treatment in older, frequently comorbid patients. For pharmacovigilance, the essential distinction is between comparative trial results, adverse reactions in a particular authorised label, and causal assessment of an individual clinical event [1–4].
- Zanubrutinib: Discovery, Development, Safety and Pharmacovigilance
- Why BTK became a therapeutic target
- Discovery, origin and early development
- Pharmacology and exposure
- The business and regulatory lifecycle
- Clinical development: what each pivotal study established
- Comparative evidence: how to read the programme
- Long-term exposure and emerging uncertainty
- Safety profile: pharmacology meets the treated population
- Drug interactions and medication errors
- Listedness: September 2026 USPI versus October 2026 EU product information
- Practical pharmacovigilance: from individual cases to signal decisions
- Signal detection, risk management and governance
- Working checklist for case reviewers
- Key takeaways
- References
- Regulatory Note
Why BTK became a therapeutic target
BTK is a cytoplasmic tyrosine kinase involved in signalling downstream of the B-cell antigen receptor and other pathways. In susceptible B-cell malignancies, this signalling helps malignant cells survive, migrate and interact with supportive tissue environments. Inhibiting BTK therefore attacks a survival pathway rather than a single histological cancer type. The diseases studied with zanubrutinib—mantle cell lymphoma (MCL), Waldenström macroglobulinaemia (WM), marginal zone lymphoma (MZL), chronic lymphocytic leukaemia/small lymphocytic lymphoma (CLL/SLL) and follicular lymphoma (FL)—are biologically and clinically distinct; efficacy in one does not establish efficacy in another [1,2].
First-generation BTK inhibition demonstrated that the target was druggable, but continuous exposure brought clinically important bleeding, infection and cardiovascular concerns. Zanubrutinib (development code BGB-3111) was designed to provide potent, selective and sustained BTK inhibition with less activity against certain other kinases than ibrutinib. Greater biochemical selectivity is a development hypothesis, not proof that off-target toxicity has been abolished. Bleeding and atrial arrhythmias remain clinically relevant with zanubrutinib [1,3,4].
Discovery, origin and early development
Zanubrutinib was discovered within BeiGene's medicinal chemistry programme and became its first internally discovered product to receive US approval. The molecule is a small-molecule, irreversible BTK inhibitor, not an antibody or a cellular therapy. It binds covalently to a cysteine residue in the BTK active site. This binding sustains target inhibition beyond the transient presence of unbound drug, although clinical exposure and tissue occupancy still depend on dose, adherence and drug interactions [1,5].
The company reported that the compound entered clinical development in 2014. The first-in-human, open-label BGB-3111-AU-003 study used dose escalation followed by disease-specific expansion cohorts. Doses explored included 40, 80, 160 and 320 mg once daily and 160 mg twice daily. In the reported dose-finding and CLL/SLL cohorts, no dose-limiting toxicities were observed during escalation. Median peripheral blood mononuclear-cell BTK occupancy exceeded 95% across studied doses, but sustained high occupancy in lymph-node biopsies was more frequent at 160 mg twice daily than at 320 mg once daily in the reported analysis (89% versus 50%). This distinction between blood and tissue pharmacodynamics was important for dose development [3].
An occupancy measurement is a pharmacodynamic biomarker; it is not a validated surrogate for survival in every lymphoma. Nor does the early phase study justify a claim that all patients must use twice-daily dosing. The current US and EU product information allows 160 mg twice daily or 320 mg once daily in the relevant labelled settings [1,2].
Pharmacology and exposure
Zanubrutinib inhibits BTK-mediated signalling and downstream processes involved in malignant B-cell proliferation, trafficking and survival. It is administered orally and metabolised primarily through CYP3A pathways. Strong and moderate CYP3A inhibitors can increase exposure; CYP3A inducers can decrease exposure and potentially undermine efficacy. A correct safety assessment therefore records not only the zanubrutinib dose but also antifungals, macrolides, antiepileptics, herbal products and the dates on which interacting medicines were started or stopped [1,2].
The approved total daily dose is generally 320 mg, taken once or in two divided doses. The US label includes 80-mg capsules and 160-mg functionally scored tablets. The EU EPAR product-information document also contains capsule and tablet presentations; product-specific instructions must be checked before advising whether a tablet may be split. Capsules must not be opened, broken or chewed. Tablets should not be crushed or chewed, but the scored tablet may be divided when directed [1,2]. A switch in presentation is not itself an adverse reaction, yet the associated change in tablet count or administration routine may cause an error.
Severe hepatic impairment requires dose reduction. Renal impairment usually does not mandate the same routine reduction, but patients with severe renal dysfunction or dialysis exposure require careful monitoring because clinical evidence is limited. Dose-modification algorithms for recurrent severe toxicity differ from ordinary missed-dose instructions; neither should be improvised from the half-life or target-occupancy data [1,2].
The business and regulatory lifecycle
The originator BeiGene subsequently adopted the BeOne Medicines name. The US DailyMed record identifies BeOne Medicines USA, Inc. as packager, and the EMA product information names BeOne Medicines Ireland Limited as marketing-authorisation holder. This is an identifiable continuity of the originator's programme and corporate branding, not evidence that the INN changed or that a new molecule was developed [1,2,5].
The US FDA granted accelerated approval in November 2019 for adults with previously treated MCL, using response data from single-arm studies. The programme then expanded into WM (August 2021), relapsed/refractory MZL after anti-CD20 treatment (September 2021), CLL/SLL (January 2023), and relapsed/refractory FL in combination with obinutuzumab (March 2024). Some US indications retain accelerated-approval language requiring confirmation of clinical benefit; a response-rate endpoint should not be described as established overall-survival benefit [1,6–9].
The European Commission authorised Brukinsa in November 2021. EU indications subsequently expanded to WM, MZL, CLL and relapsed/refractory FL with obinutuzumab, but the retrieved EU SmPC does not list the US MCL indication. This regional difference matters to benefit–risk assessment and off-label exposure classification. A US-approved use is not automatically authorised in the EU [2].
The addition of the 160-mg tablet presentation in 2025 is a meaningful lifecycle development. It may reduce pill burden, but also requires renewed attention to product identification, dispensing, tablet splitting, stock transitions and medication-error surveillance. Product presentation and local authorisation must be verified rather than assumed from a commercial announcement [1,2].
Clinical development: what each pivotal study established
A product spanning several B-cell cancers cannot be described accurately by one pooled efficacy number. Disease biology, prior therapy, response definitions and comparator choice determine what a trial can demonstrate. The following programme is best understood as a series of distinct clinical questions rather than interchangeable evidence [3,4,6–11].
Early MCL studies and the limits of single-arm response
The initial US MCL authorisation relied on BGB-3111-206, a multicentre phase II study in previously treated patients, together with a smaller MCL cohort in BGB-3111-AU-003. These studies provided evidence of tumour responses in a population with limited options. Without randomised comparison, however, the studies could not directly quantify a survival advantage over alternative therapy. This distinction explains the accelerated-approval wording retained in the US label [1,6].
For PV, a single-arm oncology study can characterise events observed under exposure, but causal attribution is difficult when the underlying cancer causes cytopenias, infection and bleeding. The timing of prior chemotherapy, marrow infiltration and concomitant treatment is as important as the study's adverse-event table.
ASPEN: Waldenström macroglobulinaemia
ASPEN was a randomised phase III study comparing zanubrutinib with ibrutinib in symptomatic WM with MYD88-mutated disease; a separate nonrandomised cohort included patients without that mutation or with unknown status. The primary endpoint in the randomised cohort was complete response or very good partial response by independent review [4,7].
In the initial publication, very good partial responses occurred in 28% of zanubrutinib recipients and 19% of ibrutinib recipients. The difference did not meet statistical significance (P=0.09), so the trial cannot be summarised as proving superiority on its primary endpoint. Atrial fibrillation, diarrhoea, oedema and some other toxicities were less frequent in the zanubrutinib arm, whereas neutropenia was more frequent. The trial did not demonstrate that serious infection or bleeding was absent with either treatment [4].
FDA's 2021 WM decision used a noncomparative analysis of response and duration of response from the zanubrutinib arms, not a claim of statistically significant superiority over ibrutinib. A regulatory indication, a randomised comparison and a postmarketing signal are three different evidentiary objects [7].
MAGNOLIA: marginal zone lymphoma
MAGNOLIA enrolled patients with relapsed or refractory MZL previously treated with an anti-CD20-based regimen. The FDA's accelerated MZL approval also considered a separate cohort from the early BGB-3111-AU-003 programme. The efficacy basis was overall response rate and duration of response in nonrandomised cohorts [1,8]. That design is informative about activity in a difficult-to-treat population but does not determine comparative progression-free survival or mortality benefit.
MZL patients often have previous rituximab, chemotherapy or corticosteroid exposure, making infection and cytopenia attribution particularly complex. Reports should document the actual regimen, treatment-free interval and prior immunosuppression rather than categorising all infections as new effects of BTK inhibition.
SEQUOIA: first-line CLL/SLL
SEQUOIA tested zanubrutinib against bendamustine plus rituximab in treatment-naive CLL/SLL without deletion 17p in the randomised cohort. A separate cohort with deletion 17p received zanubrutinib without the same randomised comparison. The FDA's initial assessment reported an independent-review progression-free-survival hazard ratio of 0.42 (95% CI 0.28–0.63) versus bendamustine/rituximab, at an estimated median PFS follow-up of 25 months [9,10].
The comparison is clinically important but not a blanket statement that zanubrutinib is preferable to all modern first-line regimens. The chemotherapy comparator, eligibility criteria and genetic cohort structure must be retained when interpreting the effect. Five-year follow-up later reported a PFS hazard ratio of 0.29 and median PFS not reached with zanubrutinib versus 44.1 months with bendamustine/rituximab, strengthening the durability of this particular comparison without proving superiority to untested alternatives [11].
ALPINE: relapsed/refractory CLL/SLL
ALPINE randomised 652 patients with previously treated CLL/SLL to zanubrutinib or ibrutinib. At median follow-up of 29.6 months, investigator-assessed progression-free survival favoured zanubrutinib (hazard ratio 0.65, 95% CI 0.49–0.86; P=0.002); estimated 24-month PFS was 78.4% versus 65.9% [12]. The study also found fewer cardiac adverse events and fewer discontinuations attributed to adverse events in the zanubrutinib arm.
These are meaningful head-to-head findings in the enrolled population, not evidence that atrial fibrillation or haemorrhage cannot occur with zanubrutinib. The endpoint hierarchy and ascertainment method matter: response, PFS, overall survival and adverse-event discontinuation are not interchangeable. A PV reviewer should not infer a universal cardiac risk ratio from one trial or extrapolate its comparison to patients with substantially different comorbidity or concomitant anticoagulation profiles.
ROSEWOOD: follicular lymphoma with obinutuzumab
ROSEWOOD randomised 217 patients with relapsed or refractory FL after at least two previous systemic treatments to zanubrutinib plus obinutuzumab or obinutuzumab alone. Independent central review found overall response rates of 69% and 46%, respectively (P=0.001). The combination received US accelerated approval in March 2024 on response and response-duration evidence [13,14].
Because obinutuzumab is itself an active anti-CD20 antibody with infusion, cytopenia and infection risks, the safety profile of this regimen must not be attributed wholly to zanubrutinib. The comparison establishes the contribution of the combination in the studied setting; it does not isolate the cause of every adverse event. Drug-specific causality may remain uncertain even when the regimen-level benefit is supported.
Comparative evidence: how to read the programme
| Study | Population and comparator | Main informative finding | Important limit |
|---|---|---|---|
| Early MCL cohorts | Previously treated MCL; single arm | Objective responses supported initial accelerated approval | No randomised survival comparison |
| ASPEN | WM; ibrutinib | Directionally higher VGPR and different toxicity pattern | Primary response difference not significant |
| MAGNOLIA | Previously treated MZL; single arm | Response and duration supported accelerated approval | Cannot quantify comparative efficacy |
| SEQUOIA | Untreated CLL/SLL without del(17p); bendamustine/rituximab | PFS benefit over specified chemoimmunotherapy | No direct comparison with all current regimens |
| ALPINE | Relapsed/refractory CLL/SLL; ibrutinib | PFS benefit and fewer cardiac events in trial | Does not eliminate residual cardiac/bleeding risks |
| ROSEWOOD | Previously treated FL; obinutuzumab alone | Higher response with combination | Combination toxicity requires dual-drug attribution |
The table also explains why a single graphic that ranks all trials by efficacy would mislead: the endpoints, diseases and comparators are not commensurate. A comparison matrix is the appropriate teaching figure for this section, so no decorative SVG is necessary.
Long-term exposure and emerging uncertainty
Continuous oral therapy can produce a different safety burden from a fixed course of chemoimmunotherapy. Adverse reactions may occur after months or years; cumulative infection exposure, cardiovascular risk and second malignancies require longitudinal follow-up. Trial participants also receive protocol-directed laboratory testing, whereas routine care may involve less frequent review, interacting medicines and treatment interruptions.
Long-term extensions and real-world data can reveal patterns under broader use but bring selection bias, missing exposure information and confounding by disease progression. A lower rate of a particular event in a randomised trial should be reported with its denominator, follow-up and comparator—not recast as a product-wide guarantee. Conversely, a spontaneous-report cluster does not establish incidence. Both forms of evidence contribute to the evolving benefit–risk assessment [1,2,4,12].
Safety profile: pharmacology meets the treated population
The US September 2026 DailyMed label reports serious haemorrhage, infections, cytopenias, second primary malignancies, cardiac arrhythmias, hepatotoxicity including drug-induced liver injury, and embryo-fetal toxicity in its warnings and precautions. These categories are clinically consequential, but their label placement does not establish that every event in a treated patient is caused by the drug. Baseline malignancy, previous therapy and concurrent medicines must remain in the analysis [1].
Haemorrhage and perioperative management
BTK inhibition can be associated with bleeding, from bruising and petechiae to gastrointestinal or intracranial haemorrhage. In the US clinical-trial safety population, grade 3 or higher haemorrhage was reported in 3.8% and fatal haemorrhage in 0.2%; these are labelled pooled clinical-trial proportions, not spontaneous-report incidence estimates [1]. Anticoagulants, antiplatelet agents, thrombocytopenia, procedures and disease-related coagulopathy can increase risk.
The US label advises discontinuation for intracranial haemorrhage of any grade and consideration of withholding treatment for three to seven days before and after surgery according to bleeding risk. The EU SmPC additionally states that warfarin or other vitamin K antagonists should not be administered concomitantly. These are product-label statements in their respective jurisdictions; a reviewer should not convert them into an identical universal policy [1,2].
For a bleeding case, collect site, severity, imaging, platelet count, coagulation studies, concomitant antithrombotics, trauma or procedure details, transfusions and dechallenge outcome. A platelet count alone cannot exclude BTK-related platelet-function effects, and a report of bruising should not be automatically coded as major haemorrhage.
Infection, viral reactivation and cytopenias
Serious bacterial, viral and fungal infections, including opportunistic infections, have occurred. The US label reports grade 3 or higher infections in 26% of the pooled clinical-trial population, with pneumonia most common, and describes HBV reactivation. The EU SmPC explicitly recommends establishing HBV status before treatment and specialist input for positive HBV serology. Both labels call for monitoring, while prophylaxis decisions depend on clinical risk and local standards [1,2].
Neutropenia, thrombocytopenia and anaemia may be due to treatment, marrow disease, previous chemotherapy or infection. Complete blood counts should be monitored during therapy; the EU SmPC specifies monthly monitoring, whereas the US label uses 'regularly'. Severe or recurrent events may require interruption or dose modification. A suspected febrile neutropenia case requires actual temperature, absolute neutrophil count, culture results, imaging, antimicrobial exposure, hospitalization and outcome [1,2].
A patient with previous HBV exposure and new hepatitis needs virological follow-up rather than a reflex diagnosis of drug-induced liver injury. Viral reactivation, biliary obstruction, sepsis, tumour infiltration and interacting medicines are relevant competing explanations for abnormal liver tests.
Arrhythmias and cardiovascular risk
Atrial fibrillation or flutter and serious ventricular arrhythmias have been reported. The US label reports atrial fibrillation/flutter in 4.4% of pooled exposed patients and grade 3 or higher ventricular arrhythmias in 0.3%. These proportions are not estimates of risk in every CLL population and cannot be directly compared with percentages from different trials [1]. Advanced age, hypertension, acute infection and structural heart disease are important modifiers.
Case review should capture ECG or telemetry confirmation, baseline rhythm, electrolytes, thyroid function when clinically relevant, infection, cardiac comorbidities, anticoagulants and the clinical consequence of arrhythmia. The decision to anticoagulate may itself create a bleeding-risk dilemma. Comparative findings in ASPEN and ALPINE support a potentially differentiated profile relative to ibrutinib in their studied populations but do not negate the labelled zanubrutinib risk [4,12].
Hepatic injury
The US label explicitly warns about potentially severe or fatal drug-induced liver injury and advises baseline and ongoing transaminase and bilirubin evaluation. Suspected DILI calls for withholding treatment, with discontinuation upon confirmation [1]. This should not be conflated with the separate dose adjustment for pre-existing severe hepatic impairment.
A medically useful hepatic case should record alanine aminotransferase, alkaline phosphatase, bilirubin, international normalised ratio, dates of change, symptoms, viral serology, imaging, concomitant medicines and dechallenge. A structured R ratio may help classify the biochemical pattern when values and laboratory upper limits of normal are available. An isolated transaminase increase is not equivalent to confirmed severe DILI. In oncology, liver infiltration, infection, obstruction and other antineoplastics can be substantial confounders.
Second malignancies, skin events and tumour lysis
Second primary malignancies, including non-melanoma skin cancers and other solid tumours, appear in the product safety information. The US label reports second primary malignancies in 14% of its pooled clinical-trial population, which is not proof that zanubrutinib caused each malignancy [1]. Age, prior immunosuppression, ultraviolet exposure, inherited susceptibility and surveillance intensity influence observed cancer diagnoses.
The EU SmPC also lists tumour lysis syndrome as an uncommon reaction and highlights its relevance in CLL, particularly with high tumour burden [2]. A tumour-lysis report should record timing after treatment initiation, tumour burden, urate, potassium, phosphate, calcium, creatinine, supportive care and competing causes. Laboratory tumour lysis and clinically significant tumour lysis are not synonymous. Skin reactions should be described morphologically; a generalised exfoliative eruption is different from uncomplicated maculopapular rash.
Pregnancy, lactation and other special populations
Animal studies support a risk of embryo-fetal harm. The US label recommends effective contraception during treatment and for one week after the last dose for patients of reproductive potential, with separate lactation advice not to breastfeed during treatment and for two weeks after the final dose [1]. Pregnancy cases need gestational exposure timing, dose, indication, prior cancer therapy, maternal course and infant outcomes; they should not be converted into a causal fetal-toxicity conclusion from temporal association alone.
Evidence in children is insufficient for routine extrapolation from adult oncology indications. Elderly patients may have higher comorbidity and polypharmacy burden. In patients with hepatic or renal impairment, the actual organ function, current label instructions and treatment interruptions should be recorded rather than relying on age alone [1,2].
Drug interactions and medication errors
The central pharmacokinetic interaction is CYP3A. Potent inhibitors such as certain azole antifungals can increase exposure, while inducers such as rifampicin can decrease exposure. The US and EU labels provide specific dose-adjustment and avoidance instructions that must be applied to the current presentation and jurisdiction; a generic instruction to halve the dose with every interacting medicine would be unsafe [1,2].
Medication errors can arise from confusing 80-mg capsules with 160-mg tablets, mixing once-daily and twice-daily regimens, incorrectly splitting or crushing a presentation, double dosing after a missed dose, or resuming the wrong dose after an inhibitor is stopped. The report should distinguish prescription, dispensing, actual administration, exposure and resulting harm. Not every error causes an adverse reaction, but errors with significant potential harm may still merit pharmacovigilance or medication-safety assessment under applicable rules.
Combination treatment creates another attribution problem. An infusion reaction following obinutuzumab, a neutropenic infection during the combination and a zanubrutinib dosing error should not be collapsed into one undifferentiated drug-event pair. Each suspect and concomitant product needs its own evidence and temporal analysis.
Listedness: September 2026 USPI versus October 2026 EU product information
Public-label comparison is essential for case triage, but it must be done against a named product and version. The US source here is the BRUKINSA capsule/tablet DailyMed label, SPL set ID 3e08fe23-d70e-424c-bc51-1222e320f902, updated 4 September 2026, whose prescribing-information highlights say revised July 2026. The EU source is the BRUKINSA EPAR product-information PDF for 80-mg hard capsules and 160-mg film-coated tablets, linked from the EMA EPAR page updated 7 October 2026. Both were accessed 26 October 2026 [1,2,16]. These version identifiers describe the retrieved public sources; the EMA page update date is not automatically the date on which every paragraph of the SmPC was first adopted.
The comparison below separates a reaction explicitly listed in adverse-reaction information from an important warning or a discussion elsewhere in the label. It is not a determination of company-core listedness, clinical-trial expectedness, SUSAR status or case-specific causality.
| Safety topic | US BRUKINSA USPI, 2026 | EU BRUKINSA SmPC, EMA page 7 Oct 2026 | Material interpretation |
|---|---|---|---|
| Bleeding/haemorrhage | Warnings §5.1; trial adverse reactions §6.1; serious and fatal bleeding described | Special warning §4.4; haemorrhage/haematoma and bruising in §4.8 | Both recognise risk; EU specifically states warfarin/vitamin K antagonists should not be coadministered |
| Infections and HBV reactivation | §5.2 describes serious/opportunistic infections and HBV reactivation | §4.4 explicitly calls for pre-treatment HBV status; §4.8 lists HBV reactivation as uncommon | Same broad concern, different explicit monitoring and frequency presentation |
| Cytopenias | §5.3 and §6.1; regular CBC monitoring | §4.4 and §4.8; monthly CBC monitoring | Do not substitute one jurisdiction's frequency wording for the other |
| Atrial fibrillation/flutter | §5.5 warning and clinical-trial reporting | §4.4 and §4.8; common frequency category | Both recognise risk; trial denominators and frequency conventions differ |
| Hepatotoxicity/DILI | Explicit §5.6 warning, including potentially fatal DILI | Retrieved SmPC does not contain an equivalent named DILI warning in §4.4 | A warning difference in assessed text, not proof that EU case assessment may disregard liver injury |
| Tumour lysis syndrome | Not named in the US warnings reviewed | Explicit §4.4 and §4.8; uncommon | EU-specific explicit presentation in assessed scope |
| Nausea | Listed in clinical-trial adverse-reaction information | §4.8 and leaflet; 2026 PRAC PSUR recommended very common in monotherapy and common with obinutuzumab | A documented EU label-evolution example, not a newly discovered clinical syndrome |
| Second primary malignancies | §5.4 warning | §4.4 warning; skin cancer highlighted | Observed diagnoses require careful confounding assessment |
| Hypersensitivity | US §4 says 'None' for formal contraindications | EU §4.3 contraindicates hypersensitivity to active substance or excipients | Formal contraindication language differs |
| Indication/presentation | US includes MCL and CLL/SLL; 80-mg capsules and 160-mg tablets | Retrieved EU SmPC includes WM, MZL, CLL and FL combination; capsule and tablet presentations | Indication and presentation scope must be checked before interpreting use or error |
What “not listed” means here. In the specifically examined EU §4.4 text, a separately headed DILI warning analogous to US §5.6 was not identified. That is a scoped comparison, not an assertion that liver abnormalities can never appear anywhere in EU product information, or that an individual hepatic event is unexpected for regulatory reporting. Likewise, the absence of a US warning heading for tumour lysis does not prove absence from every possible source of safety information. Formal listedness and expectedness decisions require the exact applicable reference safety information and event terminology, and they remain distinct from clinical causality [1,2].
The June 2026 PRAC PSUR discussion is particularly instructive. PRAC recommended adding nausea with a specified frequency to EU product information after reviewing clinical trials and solicited/spontaneous evidence, while concluding that the benefit–risk balance remained unchanged. It also requested cumulative reviews of syncope and acute kidney injury in the next PSUR. A request for further review is not equivalent to an established new adverse reaction or a mandated label change for those two outcomes [15]. The EPAR's October 2026 procedure record documents subsequent product-information updating [16].
Practical pharmacovigilance: from individual cases to signal decisions
Example 1: atrial fibrillation and anticoagulation
Illustrative, not a documented case. An older patient receiving zanubrutinib for CLL develops new atrial fibrillation after an infection and is started on an anticoagulant. A subsequent gastrointestinal bleed is reported. The safety team needs two related timelines: rhythm onset with cardiac and infection history, and bleeding onset with platelet counts, anticoagulant start date, endoscopy, transfusion and zanubrutinib interruptions. The arrhythmia and bleeding may have overlapping causes; one should not be automatically reduced to a consequence of the other.
The case narrative should state what was observed, what clinicians suspected, what the product information warns about and what remains uncertain. Coding should preserve distinct clinical concepts; seriousness and outcome must be assessed for each event. The current EU caution against warfarin/vitamin K antagonists is a separate product-use question from whether zanubrutinib caused the bleeding [1,2].
Example 2: hepatitis with multiple explanations
Illustrative. A patient develops jaundice while receiving zanubrutinib and an azole antifungal, with a history of prior hepatitis B infection. Follow-up should capture exact exposure dates, CYP3A interaction, bilirubin and liver-enzyme trends, HBV DNA and serology, imaging and disease status. A plausible DILI signal and HBV reactivation may require parallel investigation until the evidence differentiates them. If DILI is suspected, the US label recommends withholding zanubrutinib and discontinuing upon confirmation [1].
Example 3: medication error after a presentation switch
Illustrative. A prescription changes from four 80-mg capsules daily to two 160-mg tablets daily, but the patient follows both old and new instructions. The safety report should identify actual tablet/capsule counts, duration, symptoms, pharmacist counselling, packaging and whether both presentations were dispensed. The event is a dose/administration error; any bleeding, infection or other clinical outcome must be assessed separately. If a tablet was split, confirm whether the locally supplied tablet was scored and splitting was prescribed [1,2].
Example 4: post-PSUR nausea signal review
Illustrative. A periodic safety review finds more nausea reports after expansion into combination therapy. Before attributing the pattern to a new intrinsic hazard, the team should compare exposure, indication, obinutuzumab use, concomitant antiemetics, case solicitation, report completeness and time to onset. The 2026 PRAC action shows how cumulative evidence can justify a label update even when overall benefit–risk remains positive [15].
Signal detection, risk management and governance
An effective oncology PV programme connects four levels of evidence. Individual reports preserve clinical detail, chronology, seriousness, dechallenge and competing causes. Aggregate safety review examines MedDRA-coded cases, duplicates, dose and indication stratification, time to onset and exposure estimates. Comparative evidence uses trial and observational designs appropriate to the safety question, acknowledging confounding and ascertainment. Regulatory governance records the medical conclusion, reference safety information, proposed action, decision, implementation and follow-up [1,2,15].
For zanubrutinib, priority review topics include major bleeding with antithrombotics, serious infections and HBV reactivation, persistent cytopenias, atrial and ventricular arrhythmias, DILI, second malignancies, tumour lysis, severe skin reactions, medication errors and combination-treatment attribution. A signal is not established solely because a disproportionality score is elevated. Likewise, an event mentioned in a label may still warrant review if severity, risk factors, frequency or outcome changes materially.
EU GVP obligations and the applicable marketing authorisation govern case handling and periodic benefit–risk review. The use of a local case follow-up form, a cross-functional safety dashboard or a particular internal triage checklist is an operational practice unless a specific binding rule requires it. An inspection-ready record should permit reconstruction of the source, assessment, decision and action without retroactively treating a clinical hypothesis as proven.
Working checklist for case reviewers
| Domain | Minimum useful evidence | Common failure to avoid |
|---|---|---|
| Product identity | INN, presentation, strength, country, lot if available, actual regimen | Assuming capsule and tablet instructions are identical |
| Indication | Histology, stage, previous lines, mutation context where relevant | Pooling CLL, WM, MCL, MZL and FL without stratification |
| Chronology | Start/stop dates, interruptions, dose changes, interacting drugs | Assigning onset to the prescription date rather than exposure |
| Bleeding | Site, imaging, platelets, antithrombotics, procedures, treatment | Treating bruising and intracranial bleeding as equivalent |
| Infection | Organism, cultures, ANC, immune status, HBV status, prophylaxis | Missing prior immunosuppression or reactivation |
| Cardiac | ECG, rhythm history, blood pressure, infection, electrolytes | Equating a trial comparison with no residual risk |
| Hepatic | ALT, ALP, bilirubin, INR, HBV testing, imaging, other drugs | Assuming every transaminase increase is DILI |
| Follow-up | Management, dechallenge, rechallenge if any, outcome | Recording “resolved” without clinical evidence |
| Regulatory context | Dated USPI/EU SmPC, local authorisation, applicable RSI | Confusing public listedness, expectedness and causality |
Key takeaways
Zanubrutinib's development began with selective covalent BTK inhibition and tissue-occupancy studies. The clinical programme then generated distinct types of evidence: single-arm responses in MCL and MZL, a non-significant primary response comparison in ASPEN, PFS advantages in SEQUOIA and ALPINE, and a response advantage for zanubrutinib plus obinutuzumab in ROSEWOOD. The US and EU labels overlap on major risks but differ in indications, specific warnings, monitoring language and formal contraindications. The June 2026 PRAC nausea review illustrates active lifecycle pharmacovigilance without a change in overall benefit–risk. For case work, exact product identity, regimen, competing causes and the applicable dated safety reference remain indispensable.
References
- US National Library of Medicine. DailyMed: BRUKINSA (zanubrutinib) capsules and tablets, SPL set 3e08fe23-d70e-424c-bc51-1222e320f902, updated 4 September 2026 (highlights revised July 2026). Accessed 26 October 2026. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=3e08fe23-d70e-424c-bc51-1222e320f902
- European Medicines Agency. BRUKINSA: EPAR product information, SmPC §§4.1–4.8 and Annexes, linked from EPAR updated 7 October 2026. Accessed 26 October 2026. https://www.ema.europa.eu/en/documents/product-information/brukinsa-epar-product-information_en.pdf
- Tam CS, et al. Phase 1 study of the selective BTK inhibitor zanubrutinib in B-cell malignancies and safety and efficacy evaluation in CLL. Blood. 2019. https://pubmed.ncbi.nlm.nih.gov/31340982/
- Tam CS, et al. A randomized phase 3 trial of zanubrutinib vs ibrutinib in symptomatic Waldenström macroglobulinemia: the ASPEN study. Blood. 2020;136:2038–2050. https://doi.org/10.1182/blood.2020006844
- BeiGene. US FDA grants BRUKINSA accelerated approval for MCL, originator announcement filed with US Securities and Exchange Commission, 14 November 2019. https://www.sec.gov/Archives/edgar/data/1651308/000165130819000115/brukinsaapprovalpressrelea.htm
- US Food and Drug Administration. FDA approves therapy to treat relapsed/refractory mantle cell lymphoma, 14 November 2019. https://www.prnewswire.com/news-releases/fda-approves-therapy-to-treat-patients-with-relapsed-and-refractory-mantle-cell-lymphoma-supported-by-clinical-trial-results-showing-high-response-rate-of-tumor-shrinkage-300958791.html
- US Food and Drug Administration. FDA approves zanubrutinib for Waldenström's macroglobulinemia, 31 August 2021. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-zanubrutinib-waldenstroms-macroglobulinemia
- American Society of Clinical Oncology, FDA Oncology Center of Excellence notice. FDA grants accelerated approval to zanubrutinib for marginal zone lymphoma, 15 September 2021. https://www.asco.org/news-initiatives/policy-news-analysis/fda-grants-accelerated-approval-zanubrutinib-marginal-zone
- US Food and Drug Administration. FDA approves zanubrutinib for CLL/SLL, 19 January 2023. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-zanubrutinib-chronic-lymphocytic-leukemia-or-small-lymphocytic-lymphoma
- Tam CS, et al. Zanubrutinib versus bendamustine and rituximab in untreated CLL/SLL (SEQUOIA). Lancet Oncology. 2022;23:1031–1043. https://doi.org/10.1016/S1470-2045(22)00293-5
- SEQUOIA investigators. Zanubrutinib versus bendamustine and rituximab in treatment-naĂŻve CLL/SLL: median five-year follow-up. Journal of Clinical Oncology. https://doi.org/10.1200/JCO-24-02265
- Brown JR, et al. Zanubrutinib or ibrutinib in relapsed or refractory CLL. New England Journal of Medicine. 2023. https://doi.org/10.1056/NEJMoa2211582
- Zinzani PL, et al. ROSEWOOD: zanubrutinib plus obinutuzumab versus obinutuzumab in relapsed/refractory FL. Journal of Clinical Oncology. 2023;41:5107–5117. https://doi.org/10.1200/JCO.23.00775
- US Food and Drug Administration. FDA grants accelerated approval to zanubrutinib for relapsed/refractory follicular lymphoma, 7 March 2024. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-grants-accelerated-approval-zanubrutinib-relapsed-or-refractory-follicular-lymphoma
- European Medicines Agency, PRAC. Minutes of 8–11 June 2026 meeting, §6.1.8, zanubrutinib PSUSA/00010960/202511 (nausea and requested future syncope/AKI reviews). https://www.ema.europa.eu/en/documents/minutes/minutes-prac-meeting-8-11-june-2026_en.pdf
- European Medicines Agency. Brukinsa EPAR, authorisation details, product-information revision and 7 October 2026 procedural update. https://www.ema.europa.eu/en/medicines/human/EPAR/brukinsa
Regulatory Note
This is an educational synthesis, not treatment advice or a substitute for the current locally authorised product information. US and EU indications, formulations, warnings and frequency conventions may differ. Clinical-trial event rates do not establish an individual's causal relationship or spontaneous-report incidence. A publicly listed adverse reaction is not automatically the same as expectedness against the reference safety information applicable to a specific regulatory submission. Hypothetical examples are teaching scenarios, not documented safety cases.