Bortezomib

The boron atom forms a reversible covalent bond with the catalytic threonine of the beta-5 subunit of the 26S proteasome, blocking chymotrypsin-like activity. Myeloma plasma cells manufacture immunoglobulin at an extraordinary rate and depend on the proteasome to clear the resulting misfolded protein load, so they hit a terminal unfolded-protein-response stress long before normal cells do. Blocking the proteasome also stabilises IkB, which shuts down NF-kB signalling and removes a key survival and adhesion pathway in the marrow microenvironment, and stabilises pro-apoptotic proteins such as NOXA and p53. The reversibility of the boronate bond is why recovery of proteasome activity occurs within 24 hours and why dosing is intermittent.

Targets: 26S proteasome beta-5 subunit (chymotrypsin-like activity), NF-kB pathway (indirect, via IkB stabilisation), Unfolded protein response / ER stress

• · 1.3 mg/m2 per dose. Body-surface-area dosed, so a fixed microgram figure does not apply. Subcutaneous administration is now preferred over IV — the same efficacy with substantially less peripheral neuropathy.
• · 1.3 mg/m2 weekly. Widely used in older or frailer patients and in combination regimens where cumulative neuropathy is the limiting factor.

Work out your exact syringe units →

• very commonPeripheral sensory neuropathy— The defining toxicity — painful, length-dependent and sometimes permanent. Roughly 38 percent with subcutaneous dosing versus 53 percent intravenous, which is why the route changed.
• very commonThrombocytopenia— Cyclical, with a nadir around day 11 and recovery between cycles; usually does not require permanent dose reduction.
• very commonFatigue and asthenia
• very commonDiarrhoea, constipation and nausea
• commonHerpes zoster reactivation— Common enough that acyclovir or valacyclovir prophylaxis is mandatory for the duration of therapy.
• commonInjection-site reactions— Specific to the subcutaneous route; redness and induration that resolve over days.
• commonOrthostatic hypotension
• uncommonTumour lysis syndrome— Highest risk with a large tumour burden at the first cycle.
• rarePosterior reversible encephalopathy syndrome

• Intrathecal administration — it is fatal, and this has actually killed patients. Bortezomib is intravenous or subcutaneous only.
• Known hypersensitivity to bortezomib, boron or mannitol.
• Pregnancy — it is embryo-fetal toxic.

• cautionStrong CYP3A4 inhibitors such as ketoconazole or ritonavir — Raises bortezomib exposure and toxicity; monitor closely.
• conflictStrong CYP3A4 inducers such as rifampin or St John's wort — Can drop exposure enough to lose efficacy — avoid.
• redundantcarfilzomib — Both hit the same proteasome subunit; they are alternatives within a regimen, not partners.
• cautionThalidomide, lenalidomide and other neurotoxic agents — Additive peripheral neuropathy; this drives most real-world dose reductions.

• · Full blood count before each dose, watching the day-11 platelet nadir.
• · Structured neuropathy assessment at every visit — this is the toxicity that determines long-term function.
• · Blood pressure including orthostatics, particularly in dehydrated or antihypertensive-treated patients.
• · Serum paraprotein and free light chains every cycle to gauge response.
• · Antiviral prophylaxis adherence for zoster.

FDA and EMA approved, now available generically. Prescription chemotherapy administered under oncology supervision.

• Richardson et al. 2005, APEX phase 3 trial of bortezomib versus dexamethasone, New England Journal of Medicine (trial)
• San Miguel et al. 2008, VISTA phase 3 trial of bortezomib in previously untreated myeloma, New England Journal of Medicine (trial)
• Moreau et al. 2011, subcutaneous versus intravenous bortezomib phase 3 trial, Lancet Oncology (trial)
• FDA prescribing information for Velcade (label)

The boron atom is the whole story. Boronic acids are electron-deficient and form a reversible tetrahedral covalent adduct with the hydroxyl of the N-terminal threonine in the beta-5 subunit of the 20S proteasome core — the catalytic residue that performs chymotrypsin-like cleavage. That adduct dissociates on a timescale of hours, which is why proteasome activity recovers within about 24 hours of a dose and why the schedule is intermittent rather than continuous. Continuous proteasome inhibition is lethal to normal cells; the whole therapeutic window here is built out of the reversibility. Downstream, three things happen at once and it is the combination that kills the plasma cell. First, misfolded and ubiquitinated protein accumulates. A myeloma plasma cell secretes immunoglobulin at a rate that makes its endoplasmic reticulum load already near-maximal, so it hits terminal unfolded protein response stress — PERK and IRE1 signalling switch from adaptive to apoptotic, CHOP rises, and the cell commits. Normal cells with lower secretory load have slack. Second, IkB alpha is normally degraded by the proteasome to release NF-kB; stabilising IkB shuts down canonical NF-kB signalling, which in myeloma removes IL-6-driven survival signalling, adhesion molecule expression and the marrow microenvironment support that makes plasma cells so hard to kill. Third, proteasome inhibition stabilises short-lived pro-apoptotic proteins directly — NOXA in particular, which neutralises MCL-1, and p53. The neuropathy is the mechanistically interesting toxicity. Boronates are not perfectly selective; they inhibit serine proteases including HtrA2/Omi, and there is evidence for mitochondrial and endoplasmic reticulum damage in dorsal root ganglion neurons, which have a leaky blood-nerve barrier and long axons that make them uniquely exposed. The fact that subcutaneous dosing cuts neuropathy roughly in half while keeping AUC identical says the damage is driven by peak concentration, not total exposure.

The catastrophic error is route. Bortezomib given intrathecally is fatal, it has actually killed patients, and it is the reason the label carries an explicit warning. The related error is reconstitution volume — 1.4 mL for subcutaneous at 2.5 mg/mL versus 3.5 mL for intravenous at 1 mg/mL, a 2.5-fold concentration difference. Reconstituting for the wrong route and giving the concentrated solution intravenously is an overdose. The chronic error is pushing through neuropathy. It starts as numb toes, it is easy to dismiss, and by the time it is painful and length-dependent it may not fully resolve. Reduce early. The third problem is the day-11 platelet nadir being misread as marrow failure and triggering an unnecessary permanent dose reduction, when the correct response is usually to hold the dose and let the cyclical count recover. Fourth is orthostatic hypotension, which is more common than people expect, worse in dehydrated patients and those on antihypertensives, and a genuine falls risk in an elderly population with lytic bone disease. Finally, tumour lysis in the first cycle of high-burden disease — bortezomib works fast, and fast is not always kind.

1. —Cycle 1 onward — 1.3 mg/m2 subcutaneously, days 1, 4, 8 and 11 of a 21-day cycle, at least 72 hours apart. Body-surface-area dosed, so a fixed microgram value does not apply. Subcutaneous is the default route now, not a concession.
2. —On grade 1 neuropathy with pain, or grade 2 neuropathy — Reduce to 1.0 mg/m2. This is a de-escalation ladder rather than an escalation one — with this drug you start at full dose and step down, and stepping down early preserves function.
3. —On grade 2 neuropathy with pain, or grade 3 neuropathy — Hold until symptoms resolve, then restart at 0.7 mg/m2 once weekly. Many patients complete effective therapy at this dose.
4. —On grade 4 neuropathy — Discontinue. There is no dose at which grade 4 neuropathy is an acceptable trade.
5. —Alternative schedule for older or frailer patients — 1.3 mg/m2 once weekly on days 1, 8, 15 and 22 of a 35-day cycle. Widely used in transplant-ineligible patients and in combinations where cumulative neuropathy is the limiting factor, and it substantially reduces neuropathy without an obvious efficacy cost.
6. —Moderate to severe hepatic impairment, from the first dose — Start at 0.7 mg/m2 and escalate to 1.0 or reduce to 0.5 based on tolerance. This is the only baseline organ-function-driven starting-dose change.

• 26S proteasome beta-5 subunit (PSMB5, chymotrypsin-like activity) — Sub-nanomolar to low-nanomolar Ki, widely reported around 0.6 nM. Reversible covalent binding with dissociation over hours.

  Blocks the rate-limiting proteolytic step, causing accumulation of ubiquitinated protein and terminal endoplasmic reticulum stress in high-secretory cells.

• Immunoproteasome beta-5i subunit (LMP7 / PSMB8) — Also inhibited, with less selectivity between constitutive and immunoproteasome than the epoxyketones

  Contributes to activity in lymphoid malignancies and to some of the immune-modulating effects.

• NF-kB pathway, indirectly via IkB alpha stabilisation

  Removes IL-6-driven survival signalling and adhesion-mediated drug resistance in the marrow microenvironment. This is why it re-sensitises myeloma to other agents.

• Serine proteases including HtrA2/Omi — Off-target boronate reactivity

  The leading mechanistic explanation for peripheral neuropathy, and the reason the epoxyketone carfilzomib is so much less neurotoxic.

• APEX Phase 3 · n=669 · 2005

  Time to progression in relapsed myeloma versus high-dose dexamethasone. Bortezomib was superior on time to progression, response rate and one-year survival, and the trial was stopped early at interim analysis.

• VISTA Phase 3 · n=682 · 2008

  Time to progression with bortezomib plus melphalan and prednisone versus melphalan and prednisone in previously untreated, transplant-ineligible myeloma. Established the drug in front-line non-transplant disease.

• Subcutaneous versus intravenous bortezomib, randomised phase 3 non-inferiority study Phase 3, non-inferiority · n=222 · 2011

  Overall response rate after four cycles. Subcutaneous was non-inferior at 42 percent versus 42 percent, with peripheral neuropathy of any grade in 38 percent versus 53 percent and grade 3 or higher in 6 percent versus 16 percent. This trial changed the route of administration worldwide.

Proteasome inhibition is measurable within an hour of dosing and recovers within about 24 hours, which is the pharmacodynamic rhythm the whole schedule is built around. Free light chains start falling within days in light-chain disease. Intact paraprotein moves over one to two cycles, so three to six weeks. Renal function in cast nephropathy often improves within the first cycle, and that is one of the fastest and most satisfying responses in myeloma. Thrombocytopenia nadirs at day 11 and recovers by the start of the next cycle. Neuropathy is the slow one: it typically emerges over cycles 3 to 5, and cumulative dose is the main predictor. Zoster can appear at any point.

Antiviral prophylaxis is not optional. Herpes zoster reactivation is frequent enough with proteasome inhibition that aciclovir or valaciclovir runs for the whole duration of therapy and for some period afterwards, and skipping it is one of the more predictable ways to cause a preventable complication. The standard combinations are triplets and quadruplets: bortezomib with lenalidomide and dexamethasone, or with cyclophosphamide and dexamethasone, or the four-drug regimen adding daratumumab. Every one of those partners adds something bortezomib lacks, which is why it is almost never used alone. The interaction to actively avoid is a strong CYP3A4 inducer — rifampicin, carbamazepine, St John's wort — which can drop exposure enough to lose efficacy in a disease where losing efficacy means progression. Strong CYP3A4 inhibitors such as ketoconazole or ritonavir push exposure up and warrant closer monitoring rather than automatic avoidance. Stacking with other neurotoxic agents is the real-world constraint: thalidomide, vincristine and platinum agents all add to the neuropathy, and cumulative neuropathy is what ends most bortezomib courses. Green tea extract and high-dose vitamin C have both been shown in laboratory work to inactivate boronic acids directly, and while the clinical relevance is debated, there is no upside to testing it.

Against carfilzomib, the trade is neuropathy for cardiovascular risk. Bortezomib's reversible boronate causes far more peripheral neuropathy; carfilzomib's irreversible epoxyketone causes far more hypertension, heart failure and thrombotic microangiopathy. ENDEAVOR showed carfilzomib roughly doubling progression-free survival head to head at the 56 mg/m2 dose, so it is the more potent drug, but bortezomib remains the front-line workhorse because it is subcutaneous, needs no renal adjustment, is generic and is cardiac-neutral. Against ixazomib, the oral boronate, bortezomib is more potent and ixazomib is more convenient. Against daratumumab and the other anti-CD38 antibodies, these are complementary rather than competing and now routinely combined. The historical comparison worth keeping in mind is against nothing at all: myeloma median survival roughly doubled across the era this drug defined.

Not characterised here. Bortezomib is now available generically in most markets, so the acquisition price has fallen sharply from the brand era, but it is administered in an infusion centre with facility and administration charges, and it is almost always given as part of a multi-drug regimen where the other agents dominate the total. A per-vial price would not answer the question anyone is actually asking.

• ?Volume of distribution is reported in the label as 498 to 1884 L/m2, a body-surface-area-normalised range rather than an absolute litre value, so the absolute field is left null rather than converted with an assumed body surface area.
• ?The routes of elimination in humans are explicitly not characterised in the label. That is a real gap, not an omission here.
• ?The 0.6 nM Ki for the beta-5 subunit is the commonly quoted figure but I did not resolve a primary source in this session.
• ?Time to steady state is left null because clearance falls markedly between the first and subsequent doses, so conventional steady-state modelling does not describe the drug well.
• ?The HtrA2/Omi mechanism for neuropathy is one leading hypothesis among several, including mitochondrial and endoplasmic reticulum injury in dorsal root ganglia. It is not settled.
• ?APEX and VISTA enrolment figures are as commonly cited; I resolved both publication records but not participant-level detail from the full texts.
• ?The green tea polyphenol and vitamin C inactivation findings are laboratory results whose clinical significance is genuinely contested.
• ?The trial code commonly attached to the subcutaneous versus intravenous study was removed because I resolved the publication but not the sponsor study number.

• Bortezomib or high-dose dexamethasone for relapsed multiple myeloma Richardson PG, Sonneveld P, Schuster MW, et al., New England Journal of Medicine, 2005 · PMID 15958804

  APEX. The trial that established proteasome inhibition as a real therapeutic strategy rather than a laboratory curiosity.

• Bortezomib plus melphalan and prednisone for initial treatment of multiple myeloma San Miguel JF, Schlag R, Khuageva NK, et al., New England Journal of Medicine, 2008 · PMID 18753647

  VISTA. Moved the drug to front line in transplant-ineligible patients.

• Subcutaneous versus intravenous administration of bortezomib in patients with relapsed multiple myeloma: a randomised, phase 3, non-inferiority study Moreau P, Pylypenko H, Grosicki S, et al., The Lancet Oncology, 2011 · PMID 21507715

  The route-change trial. Same efficacy, roughly half the neuropathy, and a masterclass in how peak concentration rather than total exposure can drive a toxicity.

• Clinical Pharmacokinetics and Pharmacodynamics of Bortezomib Tan CRC, Abdul-Majeed S, Cael B, Barta SK, Clinical Pharmacokinetics, 2019 · PMID 29802543

  The dedicated pharmacokinetics review, useful for the clearance-drop-with-repeat-dosing behaviour that makes the terminal half-life so variable.

• VELCADE (bortezomib) for injection — US prescribing information Takeda Pharmaceuticals America, FDA prescribing information

  Source of the 20.4 versus 223 ng/mL subcutaneous-to-intravenous Cmax difference, 83 percent protein binding, 40 to 193 hour half-life, the clearance drop on repeat dosing and the CYP list.