Therapy

1. Therapeutic Goals

The goals differ sharply by disease tempo:

• Acute leukaemia (AML, ALL) — induction of complete remission (CR; <5% blasts, count recovery), consolidation to eliminate residual disease, and either further chemotherapy or allogeneic transplant for cure. Without successful induction, death within weeks.
• Chronic leukaemia (CML, CLL) — long-term disease control with minimal toxicity. Increasingly, deep responses (MMR, MR4.5, uMRD) can permit treatment-free remission.

Standard response definitions (IWG / iwCLL / ELN):

For CML the relevant terms (CHR, CCyR, MMR, MR4, MR4.5) were defined in Part IV.

2. AML — the 7+3 Induction

The cytarabine + anthracycline doublet has been the backbone of AML induction since the 1973 paper of Yates and colleagues:

• Cytarabine 100–200 mg/m²/day continuous IV infusion for 7 days — a pyrimidine analogue (ara-C) that, after intracellular phosphorylation to ara-CTP, terminates DNA chain elongation by DNA polymerase. S-phase specific.
• Daunorubicin 60–90 mg/m²/day IV push for 3 days — an anthracycline that intercalates DNA and inhibits topoisomerase II, additionally generating reactive oxygen species (the basis of cardiotoxicity, lifetime cumulative dose limit ~550 mg/m²). Idarubicin and mitoxantrone are alternatives.

CR rates: ~70–80% in patients <60 with favourable/intermediate cytogenetics; ~50% with adverse risk; ~30–50% in older fit patients. Induction mortality (death within 30 days) is ~3–5% in young patients, >15% in older patients with comorbidity. Day 14 marrow is performed to assess early response — persistent blasts trigger re-induction with 5+2 or HiDAC.

Consolidation with 2–4 cycles of high-dose cytarabine (HiDAC, 3 g/m² q12h on days 1, 3, 5 — the Mayer regimen, NEJM 1994) is the standard for favourable-risk AML. CBF AML especially benefits from HiDAC. Adverse-risk AML proceeds to allogeneic transplant in CR1 if fit.

For secondary AML / therapy-related AML, the liposomal daunorubicin/cytarabine CPX-351 (Vyxeos) showed survival benefit over standard 7+3 in older patients (Lancet et al., JCO 2018) and is now standard for these subgroups.

3. AML — Targeted Add-Ons

Several molecular subgroups now mandate addition of a targeted agent:

The differentiation syndrome (formerly “retinoic-acid syndrome”) seen with ATRA in APL also occurs with IDH inhibitors and FLT3 inhibitors: rapid proliferation of differentiating myeloid cells produces fever, dyspnoea, pulmonary infiltrates, hypoxia, weight gain — managed with high-dose dexamethasone and dose interruption.

4. AML — Elderly & Unfit Patients

The single largest therapeutic shift of the past decade. Patients >75 yr or unfit for intensive chemotherapy (ECOG 3+, severe comorbidity) historically had a median survival of 4–6 months on best supportive care. Two innovations transformed this:

• Hypomethylating agents (HMAs) — azacitidine 75 mg/m² SC d1–7 of each 28-day cycle, or decitabine 20 mg/m² IV d1–5. Modest CR rates (~20%) but improved survival vs supportive care (AZA-AML-001 Dombret, Blood 2015).
• Venetoclax (BCL2 inhibitor) added to azacitidine produces CR/CRi rates of ~65% and median OS ~14.7 months — the VIALE-A trial (DiNardo, NEJM 2020) supplanted azacitidine monotherapy as standard for newly diagnosed unfit AML.
• Glasdegib (smoothened/Hedgehog inhibitor) + low-dose ara-C is an alternative.

Venetoclax dosing requires careful TLS prophylaxis (especially in cycle 1: ramp-up from 100 → 200 → 400 mg over 3 days; allopurinol/rasburicase; daily labs). Drug interactions matter — venetoclax is a CYP3A4 substrate, so concurrent posaconazole (azole prophylaxis) requires dose reduction to 70–100 mg.

Specific subgroups within unfit AML benefit from layered targeting:

• IDH1-mutant unfit — ivosidenib + azacitidine (AGILE, NEJM 2022): OS 24 vs 7.9 mo.
• FLT3-mutant unfit — emerging triplets of HMA + venetoclax + gilteritinib (LACEWING and others).
• TP53-mutant — remains the worst prognostic subgroup; magrolimab and eprenetapopt (APR-246) trials disappointing; all-allo transplant or clinical trial.

5. APL — ATRA + Arsenic

APL is the only AML subtype routinely cured without classical cytotoxic chemotherapy. The Lo-Coco APL0406 trial (NEJM 2013; long-term Lancet Haematol 2017) randomised standard-risk APL (WBC <10×10⁹/L) to:

• ATRA + arsenic trioxide (ATO) — CR >95%, 4-yr DFS 97%, EFS 91%.
• ATRA + idarubicin (AIDA) — CR ~94%, 4-yr DFS 90%, EFS 80%.

ATRA + ATO is now the standard for standard-risk APL. High-risk APL (WBC ≥10×10⁹/L) benefits from addition of an anthracycline (gemtuzumab ozogamicin or idarubicin) or from intensified ATO regimens (Australasian Leukaemia Group).

Two acute syndromes need attention. DICdominates the first 2 weeks; aggressive replacement of fibrinogen, platelets and FFP. Differentiation syndrome(≈25% on ATRA, lower on ATO) presents with fever, weight gain, hypoxia, pulmonary infiltrates, hypotension; immediate dexamethasone 10 mg IV BID; rarely fatal if recognised early.

6. ALL — Multi-Agent Regimens

ALL therapy is the most elaborate in oncology — long, multi-phase, multi-agent regimens that take 2–3 years to complete:

• Induction (4–6 weeks) — vincristine, anthracycline (daunorubicin), corticosteroid (dexamethasone or prednisone), L-asparaginase or PEG-asparaginase, ± cyclophosphamide.
• Consolidation / intensification — high-dose methotrexate (with leucovorin rescue), cyclophosphamide, ara-C, more asparaginase, mercaptopurine.
• CNS prophylaxis — intrathecal methotrexate (or triple IT MTX/ara-C/HC), high-dose systemic methotrexate; cranial radiation now reserved for CNS-3 disease.
• Delayed intensification — repeat of induction-like block.
• Maintenance — daily 6-mercaptopurine + weekly methotrexate + monthly vincristine/steroid pulses for ~2 years total.

The German BFM (Berlin-Frankfurt-Münster)backbone underlies most international paediatric protocols (COG AALL, UKALL, DCOG, POG). Adult ALL has historically used CALGB-style regimens (10403 trial, Stock, Blood 2019, showed paediatric-style regimens work into the AYA-30s) and the hyper-CVAD regimen of MDACC (alternating cycles A and B; widely used in adult ALL but no longer first-line for younger AYA).

Asparaginase deserves special mention. Lymphoblasts cannot synthesise asparagine because they lack asparagine synthetase; extracellular depletion by bacterial asparaginase starves them of asparagine while normal cells synthesise their own. Hypersensitivity to E. coli asparaginase is managed by switching to PEG-asparaginase or Erwinia asparaginase. Pancreatitis, thrombosis (especially CVST), and hyperammonaemia are feared toxicities.

7. Ph+ ALL — TKI + Steroid Era

The Italian GIMEMA group’s Foà regimen has rewritten the management of Ph+ ALL. Early phase: dasatinib + corticosteroids alone — no cytotoxic induction — produces complete haematologic remission in ~95% within 4 weeks. Subsequent blinatumomab consolidation + dasatinib (D-ALBA trial, NEJM 2020) produced 4-year OS of 78%, MRD-negativity 60% post-induction and 81% after blinatumomab — exceeding any chemotherapy-based outcome.

The Jabbour ponatinib + blinatumomab regimen (Lancet Haematol 2024) reported similar or superior outcomes; the choice of TKI is increasingly toward ponatinib first-line because of activity against T315I (the dominant TKI-resistance mutation in ALL).

For relapsed/refractory Ph+ ALL, options include alternative TKIs, blinatumomab, inotuzumab ozogamicin, CD19 CAR-T, or allogeneic transplant. Single-cell dissection of resistance now informs sequencing decisions.

8. CML — the TKI Algorithm

The first-line choices for chronic-phase CML, with their characteristic toxicity profiles:

Choice of first-line TKI balances comorbidities: dasatinib avoided in COPD/pulmonary disease; nilotinib avoided in cardiac/diabetic patients; ponatinib reserved for T315I or multi-TKI-resistant disease (after PACE/OPTIC dose-optimisation); bosutinib often well-tolerated in older patients; imatinib still a reasonable choice for low-risk CP-CML with attention to cost.

Monitoring: BCR-ABL1^IS at 3, 6, 12 months; every 3 months in years 1–2; every 6 months thereafter once MMR is achieved. Treatment-free remission (TFR) is offered to selected patients in MR4.5 for >2 years on ELN criteria.

Blast crisis — lymphoid blast crisis is treated like Ph+ ALL with TKI + chemotherapy + transplant; myeloid blast crisis as AML with TKI; the prognosis is grave in either case.

9. CLL — BTK Inhibitors and Venetoclax

The 2014–2024 decade replaced FCR (fludarabine + cyclophosphamide + rituximab) with targeted agents in essentially all CLL settings.

• Continuous BTK inhibitor monotherapy — ibrutinib (RESONATE/-2; Burger NEJM 2015), acalabrutinib (ELEVATE-TN, Sharman Lancet 2020), zanubrutinib (SEQUOIA/ALPINE, Tam NEJM 2022; Brown NEJM 2023).
• Fixed-duration venetoclax + obinutuzumab — CLL14 trial (Fischer NEJM 2019): 12 cycles, deep MRD-negativity at end, 6-year PFS far superior to chlorambucil+obinutuzumab.
• BTKi + venetoclax doublets — CAPTIVATE (Tam, Blood 2022) and GLOW: ibrutinib + venetoclax 15 cycles, MRD-negative complete responses; FDA-approved 2024.
• Triplet — obinutuzumab + venetoclax + ibrutinib (CLL13/GAIA): not clearly superior to doublet.

Notable BTKi side-effects: atrial fibrillation (~5% on ibrutinib, less on acalabrutinib/zanubrutinib), bleeding (BTK in platelets), hypertension. Ibrutinib’s off-target effects on TEC kinases and EGFR explain most of these — acalabrutinib and zanubrutinib are more selective.

Resistance: BTK-C481S mutations (ibrutinib, acalabrutinib, zanubrutinib all bind C481); BCL2-G101V (venetoclax). Pirtobrutinib(non-covalent, BRUIN trial, FDA 2023) overcomes C481 mutations.

Practical algorithm:

• Fit, no del(17p)/TP53: venetoclax + obinutuzumab (fixed) or BTKi (continuous).
• Unfit / older: venetoclax + obinutuzumab.
• del(17p)/TP53: BTKi (continuous) or venetoclax + obinutuzumab; avoid chemoimmunotherapy.
• Relapsed / progressed on BTKi: venetoclax-based, pirtobrutinib, CAR-T, allo-SCT in select fit patients.

10. Antibody & Antibody-Drug Conjugate Therapy

Three classes are central:

• Naked monoclonal antibodies — rituximab (anti-CD20, type I), obinutuzumab (anti-CD20, type II glycoengineered), ofatumumab; mediate ADCC, CDC, and direct apoptosis. CLL workhorses.
• Antibody-drug conjugates — gemtuzumab ozogamicin (anti-CD33-calicheamicin) for AML; inotuzumab ozogamicin (anti-CD22-calicheamicin) for relapsed B-ALL (INO-VATE trial, Kantarjian, NEJM 2016: CR 80.7% vs 29.4% standard chemo). Major risk: sinusoidal obstruction syndrome (formerly VOD), particularly post-transplant.
• Bispecific T-cell engagers (BiTEs) — blinatumomab, a CD19×CD3 bispecific scFv that brings cytotoxic T-cells into contact with CD19+ B-cells. Continuous IV infusion (extremely short half-life). For MRD+ ALL it converts ~80% to MRD-negative (BLAST trial, Gökbuget Blood 2018). Major toxicities: cytokine release syndrome (CRS), neurotoxicity (ICANS-like).

Adjacent: epcoritamab (CD20×CD3) and glofitamab in relapsed B-cell lymphomas; mosunetuzumab in follicular lymphoma. The haematology/lymphoma boundary blurs further every year.

11. CAR-T Cells

Chimeric antigen-receptor T-cells are autologous T-cells genetically modified ex vivo to express a synthetic receptor combining an scFv (recognising tumour antigen), transmembrane domain, costimulatory domain (CD28 or 4-1BB), and CD3-zeta signalling chain. After re-infusion they expand, kill antigen-positive targets, and produce durable immunological memory.

FDA-approved CAR-T products in haematology (selected):

Toxicities are signature:

• Cytokine release syndrome (CRS) — fever, hypotension, hypoxia within ~3 days; IL-6 dominated; treated with tocilizumab (anti-IL-6R) ± corticosteroids; ICU support for severe cases.
• ICANS (immune-effector-cell-associated neurotoxicity syndrome) — encephalopathy, aphasia, seizures; treated with corticosteroids; tocilizumab ineffective. Higher with CD28-based products and for higher-burden disease.
• Cytopenias and B-cell aplasia — long-term B-cell aplasia from on-target CD19+ normal B-cell killing; managed with IVIg.
• Antigen-loss escape — CD19-negative ALL relapse; CD22 CAR-T or bispecific CD19/CD22 CARs in development.

For B-ALL the adoption was rapid: tisagenlecleucel produced sustained MRD-negative CR in ~80% of paediatric/young-adult patients with relapsed/refractory disease — a population that previously had <30% 5-year survival. CAR-T has now moved earlier in the algorithm: brexucabtagene for adult R/R B-ALL after first relapse; obecabtagene autoleucel (FELIX trial, NEJM 2024) approved 2024 with reduced toxicity from a fast-off CD19 binder.

12. Drug-Target Structures

Three structures encapsulate the modern leukaemia pharmacopoeia: BCR-ABL bound to imatinib (PDB 1IEP, the prototype), BTK bound to ibrutinib (PDB 5P9J, the CLL revolution), and BCL2 bound to venetoclax (PDB 6O0K, the BH3-mimetic).

Each of these crystals tells a different story: imatinib, the discovery that you can inhibit a kinase selectively in vivo; ibrutinib, the rebirth of covalent drug design; venetoclax, fragment-based discovery of a protein-protein-interaction disruptor. The corresponding three drug classes underlie the majority of haematology’s current targeted-therapy successes.