Part IV

Chronic Leukaemias — CML & CLL

Two diseases that defined the targeted-therapy era. CML — one fusion oncoprotein, one drug class, “operational cure.” CLL — the most prevalent adult leukaemia, transformed in a decade by ibrutinib, venetoclax, and the eclipse of chemoimmunotherapy.

1. The Two Chronic Leukaemias

Despite the shared adjective “chronic,” CML and CLL share almost nothing biologically. CML is a myeloid disease — a clonal expansion of granulocytic progenitors driven by a single fusion oncoprotein. CLL is a lymphoid disease — a clonal accumulation of small mature CD5+ B-cells with a much more heterogeneous molecular landscape and a slow, indolent course.

FeatureCMLCLL
LineageMyeloid (granulocyte)Lymphoid (mature B-cell)
DriverBCR::ABL1 fusion (~100%)Heterogeneous: NOTCH1, SF3B1, ATM, TP53, MYD88, BIRC3, …
Median age~64 yr~70 yr
PresentationLeucocytosis, splenomegaly, fatigue; many incidentalLymphocytosis, lymphadenopathy; many asymptomatic
Phases / stagingChronic → accelerated → blastRai 0–IV; Binet A–C
First-lineTKI (imatinib, dasatinib, nilotinib, bosutinib)BTK inhibitor (acalabrutinib, zanubrutinib) ± obinutuzumab; or venetoclax + obinutuzumab
Cure?Operational: TKI discontinuation possible in MR4.5; allo-SCT for blast crisisGenerally not curable except by allo-SCT (rare); long survival with disease control

2. CML — Ph Chromosome and BCR-ABL Biology

The Philadelphia chromosome — the small abnormal chromosome described by Peter Nowell and David Hungerford in 1960 — is the derivative chromosome 22 produced by the reciprocal translocation t(9;22)(q34;q11). Janet Rowley showed in 1973 that ABL1 (chromosome 9) is fused to BCR (chromosome 22) to produce a chimeric gene encoding a constitutively active tyrosine kinase. The breakpoints determine the isoform:

  • p210BCR-ABL (M-bcr, e13a2/e14a2) — classical adult-pattern CML.
  • p190BCR-ABL (m-bcr, e1a2) — characteristic of paediatric and adult Ph+ ALL.
  • p230BCR-ABL (μ-bcr, e19a2) — rare CML with neutrophilic phenotype.

BCR-ABL pathology depends on:

  • Constitutive kinase activity — loss of autoinhibition normally provided by the ABL N-terminal cap; oligomerisation via the BCR coiled-coil domain.
  • Ras/MAPK, PI3K/AKT, JAK/STAT5 activation — multi-pathway pro-survival, pro-proliferative signal.
  • Genomic instability — the longer a chronic-phase clone goes untreated, the more likely it acquires further mutations driving accelerated/blast phase.

The remarkable feature is that p210 BCR-ABL is sufficient. Murine bone-marrow transplantation of BCR-ABL-transduced HSCs reproduces a CML-like disease (Daley, Van Etten, Baltimore, Science 1990) — one of the cleanest single-gene cancer models in mammalian biology.

3. CML — The Three Clinical Phases

Chronic phase (CP)

~85% at diagnosis

Leucocytosis (often 50–300×10⁹/L) with full granulocytic spectrum, basophilia, thrombocytosis, splenomegaly. <10% blasts in marrow/blood. Indolent for months to years if untreated. Fatigue, abdominal fullness, sweats — or asymptomatic.

Accelerated phase (AP)

10–19% blasts

Or basophils ≥20%; persistent thrombocytopenia/cytosis; clonal evolution (additional cytogenetic abnormalities); progressive splenomegaly resistant to therapy.

Blast phase (BP)

≥20% blasts

Acute leukaemia — ~70% myeloid lineage, ~30% lymphoid. Median survival pre-imatinib ~6 months. Allogeneic transplant after re-induction is the only curative option.

Pre-1980, untreated CML progressed at ~5–7% per year through accelerated phase to blast crisis. Hydroxyurea controlled the count but did not change the trajectory. Interferon-α (1980s) gave some patients durable cytogenetic responses but with severe toxicity. The advent of TKIs has reduced annual progression to ~1–2%, and most patients diagnosed in chronic phase will never enter accelerated or blast phase.

4. The Imatinib Revolution

Imatinib mesylate (STI-571, Gleevec / Glivec) was developed at Ciba-Geigy by Nicholas Lydon and Jürg Zimmermann starting from a phenylaminopyrimidine PKC inhibitor. The compound binds the ATP-binding pocket of ABL in its inactive (DFG-out) conformation — a property that confers its selectivity for ABL, KIT, and PDGFR over the rest of the human kinome.

The pivotal trials:

  • Phase I (Druker, NEJM 2001) — 53 of 54 chronic-phase patients achieved haematologic response; many went into cytogenetic remission.
  • IRIS (NEJM 2003, NEJM 2017 long-term) — randomised imatinib 400 mg/day vs interferon + ara-C in newly diagnosed chronic-phase CML. Imatinib produced 76% complete cytogenetic response (vs 14%) at 18 months; 10-year overall survival 83%; 10-year EFS 79%; transformation to AP/BP ~6.9% in imatinib arm. The trial closed early due to crossover.
  • ENESTnd (Lancet 2010) — nilotinib 300 mg BID vs imatinib in newly diagnosed CML. Nilotinib produced higher and faster MMR rates (44% vs 22% at 12 months) and fewer progressions to AP/BP — though no overall-survival difference.
  • DASISION (Blood 2014) — dasatinib 100 mg/day vs imatinib. Dasatinib achieved earlier MMR but with more pleural effusions.
  • BFORE (J Clin Oncol 2018) — bosutinib 400 mg/day vs imatinib. Higher MMR at 12 months (47% vs 37%).

Resistance to imatinib is driven by:

  • Kinase-domain mutations — T315I (gatekeeper), Y253F/H, E255K/V, M351T, F359V, H396R; ~50–90 resistance-associated alleles described.
  • BCR-ABL amplification
  • Activation of bypass pathways (LYN, AXL)
  • Stem-cell quiescence (LSCs are largely BCR-ABL-independent)

The second-generation TKIs (dasatinib, nilotinib, bosutinib) overcome most resistance mutations but not T315I. Ponatinib — a third-generation TKI designed to accommodate the bulky isoleucine of T315I — was FDA-approved in 2012 (PACE trial, NEJM 2013) but carried significant arterial thrombotic risk; later dose-optimisation (OPTIC trial, Blood 2022) reduced this. Asciminib (FDA 2021) is the firstallosteric ABL inhibitor (binds the myristoyl pocket); ASCEMBL trial showed superiority to bosutinib in pretreated CML.

5. CML Response Milestones

CML responses are tracked by RT-qPCR for BCR-ABL1 transcripts, normalised to a control gene and reported on the International Scale (IS):

MilestoneBCR-ABL1ISEquivalent log reduction
CHR (complete haematologic response)Normal CBC; spleen non-palpable
CCyR (complete cytogenetic)~≤1%~2 log
EMR (early molecular response, 3 mo)≤10%~1 log
MMR (major molecular response)≤0.1%3 log
MR4≤0.01%4 log
MR4.5≤0.0032%4.5 log
MR5≤0.001%5 log (limit of standard qPCR)

The 2020 ELN recommendations require BCR-ABL1IS ≤10% at 3 months and ≤1% at 12 months; failure to meet these milestones triggers a switch of TKI.

Patients in deep, durable molecular response (MR4 or MR4.5 for ≥2 years) can attempt treatment-free remission (TFR). The STIM trials (Mahon, Lancet Oncol 2010; STIM2 2017) and EURO-SKI (Lancet Oncol 2018) showed ~50% of carefully selected patients remain in MMR off therapy at 3 years — the closest haematology has come to “cure” without transplant in CML, sometimes called operational cure. About half do relapse and resume TKI, almost always re-attaining response.

6. CLL — Clinical & Phenotype

CLL is the most prevalent adult leukaemia in the West (rare in East Asia). It is a clonal accumulation of small, mature, CD5+ B-cells with weak surface immunoglobulin. Diagnostic criteria (iwCLL 2018):

  • Peripheral-blood B-cell count ≥5×10⁹/L sustained ≥3 months.
  • Characteristic immunophenotype: CD5+, CD19+, CD23+, CD20dim, CD79bdim, smIgdim (κ or λ light-chain restricted), FMC7−.
  • If lymphocyte count <5×10⁹/L but lymphadenopathy present: small lymphocytic lymphoma (SLL).
  • If <5×10⁹/L without lymphadenopathy: monoclonal B-cell lymphocytosis (MBL) — a precursor with ~1%/yr progression to CLL.

Smudge cells — fragile lymphocytes that rupture during smear preparation — are a classic clue on the peripheral blood film. Most CLL patients are diagnosed incidentally on a routine CBC.

Complications of established CLL:

  • Hypogammaglobulinaemia — recurrent infections, sometimes requiring IVIg.
  • Autoimmune cytopenias — AIHA (~5–10%), ITP (~2%); sometimes precipitated by purine analogues.
  • Richter transformation — ~5–10% lifetime; transformation to diffuse large B-cell lymphoma (clonally related, aggressive) or rarely Hodgkin’s; median survival historically <1 year.
  • Second cancers — especially skin (BCC, SCC, melanoma) due to immune dysregulation and treatment effects.

7. CLL — IGHV, ZAP-70, FISH

Three molecular axes structure CLL prognostication:

  1. IGHV mutational status. Whether the variable region of the immunoglobulin heavy chain has undergone somatic hypermutation (≥2% deviation from germline). IGHV-mutated CLL behaves indolently and was historically curable with FCR (fludarabine + cyclophosphamide + rituximab). IGHV-unmutated CLL is more aggressive; relapses are typical; ibrutinib far outperforms FCR (E1912 trial, NEJM 2019).
  2. ZAP-70 expression. Aberrant expression of this T-cell signalling kinase in CLL B-cells correlates with IGHV-unmutated status and adverse outcome. Less used now than the direct IGHV sequencing.
  3. FISH abnormalities (Döhner hierarchy 2000). Five mutually-prioritised cytogenetic categories in order of worst to best:
FISH lesionFrequencyMedian TTFT (chemo era)
del(17p) / TP53~5–10% (rises with relapses)~9 months
del(11q) / ATM~15%~13 months
trisomy 12~15%~33 months
normal karyotype~20%~49 months
del(13q) sole~50%~92 months

Recurrent point mutations (TP53, NOTCH1, SF3B1, ATM, BIRC3, MYD88, POT1, XPO1, RPS15) add further granularity. TP53 mutationstatus — whether by 17p deletion or somatic mutation — predicts resistance to chemoimmunotherapy and remains the strongest single adverse predictor; these patients should never receive chemoimmunotherapy first-line and instead start with targeted therapy.

Two diseases that look like CLL but aren’t are worth distinguishing: mantle-cell lymphoma (CD5+, CD23−, cyclin-D1+, t(11;14)) and the leukaemic phase of marginal-zone lymphoma(CD5−, CD23−).

8. CLL — Rai & Binet Staging

Two clinical staging systems coexist:

Rai (USA, 1975)

  • 0: lymphocytosis only — low risk
  • I: + lymphadenopathy — intermediate
  • II: + hepato/splenomegaly — intermediate
  • III: + anaemia (<110 g/L) — high
  • IV: + thrombocytopenia (<100×10⁹/L) — high

Binet (Europe, 1981)

  • A: <3 lymphoid areas, no anaemia/thrombocytopenia
  • B: ≥3 lymphoid areas, no anaemia/thrombocytopenia
  • C: anaemia or thrombocytopenia (regardless of nodes)

Treatment is indicated for active disease per iwCLL 2018: progressive marrow failure (Rai III/IV, Binet C), rapidly progressive lymphocytosis (LDT <6 months), symptomatic / massive splenomegaly or lymphadenopathy, autoimmune cytopenia refractory to steroids, or constitutional symptoms (B-symptoms, >10% weight loss, drenching night sweats, fevers without infection).

The CLL-IPI (International Prognostic Index, 2016) integrates age >65, stage, β2-microglobulin, IGHV status, and TP53/del17p into a 4-tier score predictive of 5-year overall survival.

9. CLL — Pathway-Targeted Therapy

Three drug classes have eclipsed chemoimmunotherapy in essentially all CLL settings:

BTK inhibitors

Block Bruton tyrosine kinase, a key signal in BCR-driven survival. Ibrutinib (covalent, RESONATE/RESONATE-2, NEJM 2014/2015) — first in class. Acalabrutinib (ELEVATE-TN, ELEVATE-RR Lancet 2020) — more selective, fewer cardiac events. Zanubrutinib (ALPINE NEJM 2023) — superior PFS to ibrutinib. Resistance via BTK-C481S/F/G/Y mutations; pirtobrutinib (non-covalent, BRUIN trial) overcomes them.

BCL2 inhibitor

Venetoclax — BH3-mimetic that occupies the BH3-binding groove of BCL2, displacing pro-apoptotic BIM/BAX/BAK and triggering the intrinsic apoptosis pathway. Approved 2016 for del(17p) CLL. Combined with obinutuzumab (CLL14, NEJM 2019) gives a fixed-duration (12-month) chemo-free regimen with deep MRD-negative responses. Tumour-lysis syndrome is the major early risk; mandatory ramp-up dosing 20 → 50 → 100 → 200 → 400 mg over 5 weeks with TLS prophylaxis.

Anti-CD20 mAbs

Rituximab — type-I, original CD20 antibody, often combined with FCR or BR; the foundation of chemoimmunotherapy. Obinutuzumab — type-II, glycoengineered for enhanced ADCC; deeper responses than rituximab in CLL11 trial, now standard partner with venetoclax in CLL14.

PI3K inhibitors (less used)

Idelalisib (PI3Kδ) — effective but burdened with serious autoimmune-like toxicity (colitis, pneumonitis, hepatitis); largely supplanted. Duvelisib (PI3Kδγ) and copanlisib (pan-PI3K) remain in adjacent indications.

In 2025 the question is no longer whether to use targeted agents but which and for how long: continuous BTK inhibitor (BTKi monotherapy or BTKi + venetoclax) vs fixed-duration venetoclax + obinutuzumab vs the emerging BTKi + venetoclax + anti-CD20 triplet. MRD-guided de-escalation (FLAIR, GLOW, CAPTIVATE) is reshaping the algorithm.

10. Drug-Target Structures

Two crystals define modern chronic-leukaemia therapy: the BTK kinase domain bound to ibrutinib (PDB 5P9J), and ABL kinase — this time bound to the second-generation TKI dasatinib in active conformation rather than imatinib’s DFG-out (PDB 2GQG).

BTK kinase + ibrutinib (5P9J)

Bruton tyrosine kinase (BTK) covalently bound by ibrutinib at Cys481. The Michael-acceptor acrylamide of ibrutinib forms an irreversible bond — explaining once-daily dosing and BTK-C481S resistance.

Drag to rotate · scroll to zoom · right-drag to pan. Powered by 3Dmol.js (Rego & Koes 2014).

ABL kinase + dasatinib (2GQG)

ABL kinase bound to dasatinib in the active (DFG-in) conformation. Unlike imatinib (DFG-out), dasatinib occupies the active conformation — explaining its activity against most imatinib-resistant ABL mutations except T315I.

Drag to rotate · scroll to zoom · right-drag to pan. Powered by 3Dmol.js (Rego & Koes 2014).

Venetoclax binding to BCL2 (PDB 6O0K) is covered in Part VII, where we walk through the BH3-mimetic rationale and the FBDD campaign that led from ABT-737 to navitoclax to venetoclax.

Key references for further reading. Druker et al., Five-year follow-up of imatinib in CML (IRIS), NEJM 2006; Hochhaus et al., Long-term outcomes of imatinib in CML (IRIS 10-yr), NEJM 2017; Saglio et al., Nilotinib vs imatinib (ENESTnd), NEJM 2010; Mahon et al., Discontinuation of imatinib in CML (STIM), Lancet Oncol 2010; Hochhaus et al., ELN recommendations for CML, Leukemia 2020; Hallek et al., iwCLL guidelines for diagnosis and treatment of CLL, Blood 2018; Burger et al., Ibrutinib as initial therapy in CLL (RESONATE-2), NEJM 2015; Fischer et al., Venetoclax + obinutuzumab in CLL (CLL14), NEJM 2019; Brown et al., Zanubrutinib vs ibrutinib in CLL (ALPINE), NEJM 2023.
Share:XRedditLinkedIn