Timing and trajectory of BCR::ABL1-driven chronic myeloid leukaemia

Kamizela, Aleksandra E.; Leongamornlert, Daniel; Williams, Nicholas; Wang, Xin; Nyamondo, Kudzai; Dawson, Kevin; Spencer Chapman, Michael; Guo, Jing; Lee, Joe; Mane, Karim; Milne, Kate; Green, Anthony R.; Chevassut, Timothy; Campbell, Peter J.; Ellinor, Patrick T.; Huntly, Brian J. P.; Baxter, E. Joanna; Nangalia, Jyoti

doi:10.1038/s41586-025-08817-2

Download PDF

Article
Open access
Published: 09 April 2025

Timing and trajectory of BCR::ABL1-driven chronic myeloid leukaemia

Nature volume 640, pages 982–990 (2025)Cite this article

22k Accesses
3 Citations
115 Altmetric
Metrics details

Subjects

Abstract

Mutation of some genes drives uncontrolled cell proliferation and cancer. The Philadelphia chromosome in chronic myeloid leukaemia (CML) provided the very first such genetic link to cancer^1,2. However, little is known about the trajectory to CML, the rate of BCR::ABL1 clonal expansion and how this affects disease. Using whole-genome sequencing of 1,013 haematopoietic colonies from nine patients with CML aged 22 to 81 years, we reconstruct phylogenetic trees of haematopoiesis. Intronic breaks in BCR and ABL1 were not always observed, and out-of-frame exonic breakpoints in BCR, requiring exon skipping to derive BCR::ABL1, were also noted. Apart from ASXL1 and RUNX1 mutations, extra myeloid gene mutations were mostly present in wild-type cells. We inferred explosive growth attributed to BCR::ABL1 commencing 3–14 years (confidence interval 2–16 years) before diagnosis, with annual growth rates exceeding 70,000% per year. Mutation accumulation was higher in BCR::ABL1 cells with shorter telomere lengths, reflecting their excessive cell divisions. Clonal expansion rates inversely correlated with the time to diagnosis. BCR::ABL1 in the general population mirrored CML incidence, and advanced and/or blast phase CML was characterized by subsequent genomic evolution. These data highlight the oncogenic potency of BCR::ABL1 fusion and contrast with the slow and sequential clonal trajectories of most cancers.

Minimal residual disease in BCR::ABL1-positive acute lymphoblastic leukemia: different significance in typical ALL and in CML-like disease

Article 06 August 2022

The structure of the hematopoietic system can explain chronic myeloid leukemia progression

Article Open access 03 April 2023

Continuous therapy response references for BCR::ABL1 monitoring in pediatric chronic myeloid leukemia

Article Open access 24 October 2023

Main

Chronic myeloid leukaemia (CML) occupies a landmark position in the history of oncology research, marking the first instance in which a genetic anomaly was implicated in the development of cancer. The seminal discovery of the Philadelphia (Ph) chromosome in 1960 by Nowell and Hungerford¹ and BCR::ABL1 fusion gene in 1973 by Rowley², heralded the era of oncogenomics. Targeting BCR::ABL1 by means of tyrosine kinase inhibition (TKI) has since resulted in uniquely successful patient outcomes in CML, a result not replicated for most other cancers³.

Cancers emerge from the stepwise accumulation of key genetic mutations critical to cell growth and regulation⁴. Such mutations accumulate over an extended period, commencing decades before clinical presentation, for example, early in life whole-genome duplication in ovarian cancer and chromosome 3p loss in clear cell renal cell carcinoma^5,6. Cancer evolution may even commence in utero, as demonstrated for JAK2 mutations in adult-onset polycythaemia vera⁷. By contrast, cancer incidences in Japanese survivors of the atomic bombs showed a peak in CML within 10 years of radiation exposure, raising the possibility that BCR::ABL1 driven clonal expansion and the trajectory to CML are unlike that of adult malignancies studied so far⁸.

Somatic mutations accumulate in haematopoietic cells throughout life in a clock-like fashion^7,9. The resulting unique mutational profiles of individual cells can be harnessed to reconstruct phylogenetic trees that depict ancestral cellular relationships and evolutionary history. This approach has enabled precise quantification of clonal dynamics in both healthy haematopoiesis and haematological malignancy^7,9,10. Here, using genome-wide somatic mutations and phylogenetic inference, we characterize the fitness and trajectory of BCR::ABL1 driven clonal expansion and how these factors affect clinical features of CML.

Driver mutations in CML colony genomes

We studied nine patients, aged 22–81 years at presentation, with chronic phase CML (Fig. 1a and Supplementary Table 1). Patients harboured a BCR::ABL1 translocation t(9;22)(q34;q11) with typical fusion transcripts e14a2 or e13a2 involving the major breakpoint cluster region of BCR. Patients had varying therapeutic responses to TKI therapy. Three patients were responsive to first-line Imatinib or Dasatinib, six required second line TKI and two patients failed several different TKIs (Supplementary Table 1 and Extended Data Fig. 1). Blood and bone marrow were sampled from diagnosis in six out of nine patients and further during therapy in four patients (Fig. 1a). In three individuals (PD57333, PD57334 and PD57335), sampling time points were only after diagnosis (7 months to 2 years and 9 months, Fig. 1a and Supplementary Table 1). Mononuclear cells (MNCs) were cultured in vitro to provide single-cell derived clonal DNA for whole-genome sequencing (WGS) (Fig. 1b). Following quality control, 179 out of 1,013 colonies were excluded due to low sequencing coverage or reduced clonality. In total, 834 whole genomes were taken forward (mean 93 whole genomes per patient (range 35–163), mean depth of sequencing 15.3 times (range 8.6–46.3 times)). A total of 397,063 autosomal single-nucleotide variants (SNVs) and 13,646 autosomal insertion and deletions were identified.

**Fig. 1: Patient cohort and study design.**

We detected BCR::ABL1 in eight out of nine individuals (Fig. 1c–e). In PD57333, sampled at molecularly detectable relapse (BCR::ABL1/ABL1 ratio of 3%), we only captured BCR::ABL1 in whole blood WGS (two DNA read pairs in average read depth 45.9×), but not from the cultured colonies due to the low mutant clonal fraction. WGS provides a unique opportunity to interrogate the BCR::ABL1 breakpoint due to the ability to reconstruct exact genomic BCR::ABL1 (derivative, der(22)) and ABL1::BCR (der(9)) breakpoints, not otherwise accessible from routine clinical complementary DNA (cDNA) analysis. Two observations were apparent. First, deleted or duplicated DNA regions (length 1–687 base pairs (bp) across der(22) and der(9)) were observed at the point of fusion (Supplementary Table 2). In two patients, this completely disrupted der(9). In PD51635, there was a complex three-way translocation resulting in der(22), but also t(9;17)(q34.12;q25.3) and t(17;22)(q25.3;q11.23) resulting in an atypical der(9) (Fig. 1d and Supplementary Table 2). In PD51633, we observed large roughly 1 Mb losses deleting both 5′ ABL1 and 3′ BCR along with 55 extra genes (Fig. 1e and Supplementary Table 2). We also observed non-templated inserted sequences at points of fusion (Extended Data Fig. 2b,c). Second, whereas several individuals harboured the expected intronic breakpoints within BCR and ABL1, we observed three cases (PD56961, PD57332 and PD57335) with exonic breakpoints in BCR, an occurrence only rarely reported in the literature^11,12. All three cases predicted out-of-frame fusions with ABL1, and continuation of the fusion reading frame would result in a stop codon within 1–17 codons (Extended Data Fig. 2a–c). Nevertheless, these patients harboured typical clinically detectable fusion transcripts (Extended Data Fig. 2d). These fusion transcripts would have only been compatible with the DNA derived event if the out-of-frame exon was spliced out. Using SpliceAI, all cases showed a reduced splice acceptor probability for the out-of-frame exon (Δ score PD56961 0.49, PD57332 0.67 and PD57335 0.75), supporting this hypothesis (Extended Data Fig. 2e).

Within BCR::ABL1-positive colonies, we found occasional further mutations in ASXL1 (L983fs*8, E635fs*15) in PD51635 and PD51634. In PD57332, a canonical RUNX1 RUNT domain truncating mutation was present as a dominant subclone after BCR::ABL1 acquisition (Figs. 1f and 2). These driver mutations have been reported to affect patient prognosis, treatment response and disease transformation^13,14,15,16. Further driver mutations were more commonly found in BCR::ABL1-negative colonies, particularly trisomy 8 (six events), DNMT3A mutations (14 events) and loss of Y, but also mutations in TET2, NF1, CUX1, STAG2, PHF6, BCR, RB1 and TP53 (Figs. 1f and 2). Perhaps, it is not surprising that mutations in these genes are not consistently associated with CML features and outcomes^15,16,17, given that they are not preferentially observed within the CML clone.

**Fig. 2: Phylogenetic trees of chronic phase CML.**

Phylogenetic trees in patients with CML

Phylogenetic trees of haematopoietic cells in nine patients with CML are depicted in Fig. 2, and show the pattern of sharing of genome-wide somatic mutations across individual colonies. A shared branch represents mutations identified in downstream descendant lineages, and mutations within the lowest branches are only found in single colonies. Several general observations can be made. As would be expected, the total number of mutations in individual colonies increases with patient age, in keeping with clock-like mutation acquisition in human haematopoietic stem and progenitor cells (HSPCs)^9,10,18. All trees have an abundance of coalescences (branching points) near the root, or ‘top’, of the trees. This reflects the rapid expansion of HSPCs and resulting genomic divergence during embryogenesis, such that for any two normal HSPCs, their most recent common ancestor (MRCA) is near the start of life. Following early expansion, normal haematopoiesis in young patients have a ‘comb’-like phylogenetic structure, whereas older patients have many clonal expansions⁹. We observed prominent intra-patient recurrence for specific driver mutations, for example, eight independent DNMT3A mutations in PD51635, three independent loss of chromosome Y events in PD57333, two independent chromosome 3 aberrations in PD51633 and recurrent trisomy 8 in PD51634, strongly suggesting the existence of patient-specific selection landscapes.

A striking feature of these trees is the pattern of branching in BCR::ABL1-positive colonies. These colonies emerge below a long, vertical branch that only carries one identified driver among all somatic mutations assigned to this branch: the BCR::ABL1 translocation. The BCR::ABL1-positive clade in six patients has a rapid burst of coalescences emerging directly beneath this shared branch, akin to that observed at the start of life, indicating rapid division into a large pool of leukaemic HSPCs. In two patients (PD51635 and PD51633), a relatively slower, but still recent clonal expansion pattern is observed. These data indicate that the MRCA and commencement of tumour clonal expansion in patients with CML are recent events, contrasting with patterns observed for other driver mutations in haematopoiesis^5,6,7,9.

Mutation accumulation in BCR::ABL1 cells

Mutations accumulated in BCR::ABL1-negative HSPCs at 18 mutations per year (95% bootstrapped confidence interval, CI_boot, 16.1–20.2) in line with published data^7,9. BCR::ABL1-positive HSPCs had roughly 90 extra mutations (mean +91.7, CI_boot 33.6–151.6, Fig. 3a and Supplementary Note 1). Indeed, in many patients, the mutation burden of BCR::ABL1-positive colonies was equivalent or higher than BCR::ABL1-negative HSPCs despite being sampled at a younger age (PD51634 and PD51632) (Figs. 2 and 3a).

We considered whether these increased mutations were due to new or existing mutational processes. In BCR::ABL1-positive and negative colonies, expected endogenous clock-like mutational processes (single-base substitution (SBS) signatures SBS1 and SBSblood)^9,18,19 were present in addition to SBS18 (Fig. 3b,c). No additional mutational processes were found, contrary to previous reports of BCR::ABL1-induced DNA damage²⁰. We found no evidence of TKI-induced mutations, however, recurrent trisomy 8 was observed in BCR::ABL1-negative colonies (Figs. 1f and 2), which could reflect increased proliferation or survival of trisomy 8 HSPCs during CML development, or their subsequent positive selection by TKI.

To explore how existing mutational processes contribute to increased mutation burden in BCR::ABL1-positive cells, we assigned mutational signatures to individual branches of phylogenies (Fig. 3b and Extended Data Figs. 3 and 4), grouping by (1) BCR::ABL1-negative (wild-type) branches, (2) shared branch of the BCR::ABL1 clade representing the pre-CML lineage, (3) early shared branches of the BCR::ABL1 clade, representing historical early CML clonal expansion, (4) all branches of the BCR::ABL1 clade, including recent mutations and (5) early-in-life branches. An increased proportion of SBS1 mutations was observed in the BCR::ABL1 clade (Fig. 3c and Extended Data Fig. 3, +13.6–30.9%, P = 4.0 × 10⁻¹¹), mirrored by an increased proportion of C>T transitions at CpGs (Fig. 3d and Extended Data Fig. 4, +16.6–31.3%, P = 2.9 × 10⁻¹¹). This reflects an increased spontaneous deamination of methylated cytosines resulting from increased cell divisions²¹ (Fig. 3e). A similar pattern was observed for SBS18, which has also been observed in the rapid growth contexts of fetal haematopoiesis¹⁹ and placental tissue²². Overall, our data show that increased mutations in CML cells simply reflect their higher rate of cell division.

We observed no differences in mutational patterns between the pre-CML lineage (shared branch of the BCR::ABL1 clade) and wild-type HSPCs (Fig. 3e). This confirms that the CML cell-of-origin acquired mutations similarly to normal HSPCs before its transformation by BCR::ABL1, which probably occurred at the end of the long shared branch.

It was unusual to sample BCR::ABL1-positive colonies years after diagnosis, as most patients were in molecular remission. PD51635 had clinically stable and low BCR::ABL1/ABL1 ratios (0.5%) on TKI (Extended Data Fig. 1), and their two sampled BCR::ABL1-positive colonies (Fig. 2) had a lower proportion of C>T at CpGs (mean 0.18) when compared with BCR::ABL1-positive genomes at diagnosis (mean 0.26, Fig. 3f) suggesting that these lineages were dividing at normal rates despite harbouring BCR::ABL1. We confirmed that the BCR::ABL1 translocation events in these colonies and the diagnostic sample were identical. This unusual observation raises the possibilities that TKI therapy resulted in preferential clearance of only more rapidly dividing BCR::ABL1-mutated cells, that TKI therapy slows cell division in some BCR::ABL1-positive cells or that the downstream consequences of BCR::ABL1 have been epigenetically silenced in this patient. Competing clonal haematopoiesis or immune regulation may also have constrained cell division rates of remaining BCR::ABL1 cells.

Telomere lengths in CML

Telomeres, the repetitive DNA sequences at chromosome ends that shorten with cell division, are a direct read out of cell division history. In healthy HSPCs, telomere attrition has been shown to be roughly 30 bp per year⁹, similar to that observed here in wild-type cells (n = 469, −30.8 bp per year, CI_boot −12.6–48.2 bp). The model intercept, which provides an estimated telomere length at birth, was 6,180 bp (CI_boot 5,216–7,086 bp) in line with previous estimates⁹. BCR::ABL1-positive HSPCs (n = 365) had a marked decrease in telomere length despite being sampled at a younger age, with an extra 556.9 bp of loss independent of age (CI −86.0–993.2 bp, Supplementary Note 2 and Fig. 3g). We did note the unusually short telomeres in PD51635 BCR::ABL1-negative HSPCs, which could reflect the increased cell division history of clonal haematopoiesis within the wild-type compartment (Fig. 3h).

Timing and rate of BCR::ABL1 expansion

We calculated the timing of the start of BCR::ABL1 clonal expansion by modelling mutation accumulation at a constant rate until the commencement of clonal expansion, following which there was a higher rate of mutation acquisition, as shown above. The timing of the coalescences is inferred using our Markov chain Monte Carlo method rtreefit²³ adapted to account for the transformation event occurring at the end of the long ‘trunk’ preceding the clonal expansion (Supplementary Note 3). With these ‘time-based’ trees, we next estimated growth rates using our phylofit⁷ tool, which assumes a sigmoidal clonal outgrowth curve. The upper end of the prior range of exponential phase growth rate, s was increased until the posterior estimate of s was insensitive to further increases (Extended Data Fig. 5a). We report annualized growth rate, S = e^s − 1, as a percentage, alongside instantaneous per year growth rate, s, where in a small time period, δt, measured in years, the fractional increase in the clone size is roughly sδt.

In the two youngest patients (PD57334, 22 years; PD56961, 36 years), the time from onset of clonal expansion to clinical diagnosis was only 3.2 years (CI 2.4–4.2 years) and 3.9 years (CI 3–5.3 years, Fig. 4a and Supplementary Tables 3 and 5). In keeping with recent tumour origins, estimated growth rates were very high at 99,000,000% per year (CI 906,000–13,500,000,000% per year, s = 13.81, CI 9.11–18.72) and 26,800% per year (CI 11,800–63,000% per year, s = 5.59, CI 4.78–6.44, Fig. 4b and Supplementary Tables 3 and 5). These translate to the mutant clade doubling in size every 18 to 45 days, respectively. Three middle-aged patients (45–55 years, PD51632, PD51634, PD57335) also had recent and rapid BCR::ABL1 clonal expansion to diagnosis, ranging from 4.4 years (CI 3.4–5.8 years) with a growth rate of 73,500% per year (CI 24,000–240,000% per year, s = 6.60 CI 5.49–7.78, PD57335) to 6.3 year (CI 5.0–8 years) with a growth rate of 2,700% per year (CI 1,700–4,300% per year, s = 3.33 CI 2.89–3.79, PD51634) (Fig. 4a,b and Supplementary Tables 3 and 5).

**Fig. 4: Timing and fitness of *BCR::ABL1-*driven clonal expansion.**

By contrast, in three patients aged 52 years (PD51633), 61 years (PD51635) and 81 years (PD57332), we observed slightly longer clonal trajectories of 13.1 years (CI 10.5–16 years), 14.0 years (CI 12.0–16.2 years) and 10.3 years (CI 7.9–13.4 years), respectively. These patients had intermediate rates of clonal expansion, nonetheless, achieving growth at 114–245% per year (s = 0.76–1.24, Fig. 4a,b). Of note, the growth rate in PD51635 seemed initially to be slower (as judged by longer inter-coalescent intervals) and then more rapid within a subclone (Fig. 2 and Supplementary Tables 3 and 5).

Given such extremely rapid inferred growth rates, we benchmarked phylofit to assess whether large growth rates could be accurately inferred (Supplementary Note 3). Orthogonal estimates of growth rate using cloneRate²⁴ (‘maxLikelihood’ and ‘birthDeathMCMC’) gave similar growth estimates (Extended Data Fig. 5b and Supplementary Table 4), providing extra support for the very high estimated growth rates in CML. A notable exception is PD57335 in which late coalescences in the mutant clade have a noticeable impact on cloneRate estimates but not when using phylofit (Supplementary Table 4). As previously reported²⁴, phylofit credibility intervals were less conservative than those from cloneRate.

Acknowledging the small number of patients, there seemed to be a plausible trend of younger age of onset correlating with more explosive growth together with a shorter duration between the beginning of BCR::ABL1 clonal expansion and diagnosis (Fig. 4a,b), and older individuals demonstrating relatively less explosive clonal expansion. PD51632 was a slight outlier to these trends, having had treatment for colorectal cancer with chemotherapy in the preceding 8 months before CML diagnosis, which may have enhanced selection on BCR::ABL1-positive HSPCs.

We explored the clinical significance of differences in CML growth rates and clinical response. All three of the fastest growing CMLs in the cohort failed to achieve optimal molecular remission (defined as BCR-ABL1/ABL1 ratio less than 1% by international standards) at 12 months due to TKI-induced cytopenias and slow reduction in BCR::ABL1 levels (Extended Data Fig. 1). By contrast, two of the slowest growing CML clones in the oldest patients achieved rapid major molecular remissions. These observations raise the possibility that growth rates in CML may affect early therapeutic response.

Despite the varying clinical presentation patterns within this small cohort, we observed high correlation between the inferred growth rates of CML and the time to clinical presentation (Fig. 4c,d). Given that these two parameters were inferred from distinct features of the tree, namely, patient age at the start of the clonal expansion versus the pattern of coalescences within the clonal expansion (Fig. 2 and Methods), this pattern was striking. This observation provides evidence for the predictability of clinical presentation based on tumour growth rates and suggests that there is a limit to the population size of CML leukaemic stem cells that can be tolerated before inevitable clinical presentation (Supplementary Note 4).

BCR:ABL1 and CML in the All of Us cohort

The relative timing of mutations within a branch of a phylogenetic tree cannot be disentangled. Although we can accurately estimate the timing of commencement of BCR::ABL1 clonal expansion, we cannot pinpoint exactly when BCR::ABL1 occurs along the preceding long shared branch. Circumstantial evidence strongly suggests that BCR::ABL1 occurs at the end of this branch, triggering clonal expansion: (1) mutational patterns in the BCR::ABL1 branch are identical to wild-type HSPCs, suggesting absence of BCR::ABL1 until the end of the branch; (2) BCR::ABL1 is sufficient to induce a CML-like phenotype in mice²⁵, indicating that its acquisition triggers transformation; (3) no additional recurrent genetic driver mutations were found on shared branches harbouring BCR::ABL1; (4) pre-existing clonal haematopoiesis was not observed upstream of BCR::ABL1 and (5) acquiring BCR::ABL1 earlier in the branch would necessitate additional non-genetic events to consistently trigger clonal expansion in every patient, making this explanation less parsimonious and violating Occam’s razor.

Nevertheless, by only studying patients with CML, any possibility that BCR::ABL1 may also drive slower, decades long expansion, such as CH, may be overlooked. Whereas BCR::ABL1 has not been reported in CH, such studies have generally only focussed on point mutations, insertions or deletions or chromosomal copy number changes^26,27. Furthermore, studies of genomic translocations in healthy blood have typically used whole blood cDNA and not assessed for the presence of stable HSPC clones²⁸. To explore this, we examined the US-based All of Us cohort²⁹ of 206,173 whole-genome sequenced participants with linked electronic health data for evidence of BCR::ABL1 CH. We identified 39 individuals (0.019%) with two or more supporting reads of BCR::ABL1 or ABL1::BCR, most of whom had a diagnosis of CML, haematological features suspicious of CML (for example, splenomegaly or basophilia) or unspecified haematological issues (Extended Data Fig. 6a). Although 14 individuals did not have a CML-relevant status reported, of these, three had blood sampling after their most recent clinical record and three had no clinical information beyond 1 year after sampling (Extended Data Fig. 6b). All those with follow-up data exceeding 2.5 years had a CML-relevant status reported. These findings suggest that a readily detectable BCR::ABL1 clone in the healthy population, in the absence of a CML diagnosis, is genuinely rare. In fact, the prevalence of BCR::ABL1 in the All of Us (0.019%) closely matches the prevalence of CML in the United States (0.02%)³⁰, indicating that BCR::ABL1 does not also drive asymptomatic CH.

We also identified 40 individuals (0.019%) with a single read of BCR::ABL1 or ABL1::BCR. The confidence in these cases is lower, and more than half lacked features of CML. We hypothesized that this group could include genuine patients with CML (for example, prediagnosis or treated), BCR::ABL1 artefacts, or BCR::ABL1 in non-HSPCs. To address the possibility of sequencing artefacts, we screened roughly 10,000 HSPC whole genomes from single-cell derived haematopoietic colonies, previously sequenced at the Sanger Institute^7,9,19,31,32, for BCR::ABL1. Given the clonality of these genomes, bona fide mutations should have a variant allele frequency (VAF) of 50%. WGS coverage was 10–15× and no donors were known to have CML. Across roughly 10,000 colonies, only three reads, each from a different colony, were translocated between BCR and ABL1 target regions, but none were in the correct orientation for BCR::ABL1 or ABL1::BCR. This suggests that single BCR::ABL1 reads in All of Us are less likely to be sequencing artefacts, raising the possibility that some of these cases represent mutated, differentiated blood cells that cannot maintain clones over time. If this were the case, we would not expect such clones to increase in frequency with age. Indeed, single read BCR::ABL1 carriers showed a more uniform incidence across age, whereas carriers with two or more reads, as well as patients with CML, showed increased incidence with age (Extended Data Fig. 6c). Overall, our analyses demonstrate that BCR::ABL1 does not commonly cause CH, supporting the one-hit model of CML model depicted by our phylogenetic data.

Disease progression in CML

Progression to blast phase CML is associated with extra mutations in RUNX1, BCOR, TP53, ABL1, +8 and isochromosome 17q (i(17q))^17,33,34. Paired-exome and targeted gene sequencing before and after CML progression have also shown increased mutation burden^15,35. Using WGS of peripheral blood MNC or granulocyte DNA, we analysed four patients (aged 38–81 years) with advanced CML, characterized by elevated circulating blast cells (10–19%). In all cases, we observed further driver mutations within the BCR::ABL1 clone, such as, RUNX1p.Arg166*, ASXL1p.Tyr59fs, ASXL1.p.G646fs*12 and BCORp.Leu532fs, along with copy number alterations such as +1q, −16q, +8 and i(17q) (Supplementary Table 6). BCR::ABL1 with additional drivers expanded to dominate the CML clone, leaving little evidence of the historical BCR::ABL1 ‘only’ clone. This contrasts with chronic phase CML in this study, in which additional drivers within the BCR::ABL1 clone were either absent, or at low levels, in most cases (six out of eight). In two chronic phase cases, we did observe high clonal burden for additional driver mutations: in utero NF1 frameshift (PD51634), and RUNX1 mutation (PD57332). PD51634 subsequently transformed to blast phase CML, whereas PD57332 died soon after treatment initiation, with uncertainty on the phase of the underlying disease. These data confirm that advanced CML is largely genomically driven, characterized by BCR::ABL1 clones that acquire additional driver mutations that offer further selection advantage.

One patient with advanced-phase CML (PD60243) developed blast crisis. The 81-year-old patient had elevated blasts (11% CD7⁺/CD13⁺/CD33⁺/CD117⁺/HLA-DR⁺ cells) at presentation with RUNX1 p.Arg166* and DNMT3A p.Ala571_Ala572del mutations in addition to BCR::ABL1 (Fig. 4e,f). This clone shared at least 1,259 mutations, not dissimilar to the expected number of mutations for normal HSPCs at this age, suggesting that the MRCA of this clone arose relatively recently (probably in the seventh decade of life). Two subclones were detected: (1) trisomy 8 (56% of BCR::ABL1/RUNX1/DNMT3A-positive cells) and (2) in-frame PHF6p.H329delH (14% of BCR::ABL1/RUNX1/DNMT3A cells) (Fig. 4e,f). A few months after treatment with Nilotinib, the patient developed roughly 20% blasts in the peripheral blood, at which point the PHF6-mutated subclone was clonally dominant. This clone showed further focal events—BTG1 deletion (chromosome 12), rearranged TRB (chromosome 7) and rearranged IgH (chromosome 14)—typical changes of B cells consistent with lymphoid blast crisis. This confirms that lymphoid identity and differentiation emerged after BCR::ABL1 acquisition by means of RAG-mediated genomic evolution. There were five different ABL1-mutated subclones within this highly mutated BCR::ABL1 clone, with ABL1.p.Phe359Val as the dominant mutation (65% of cells). Other ABL1 mutations included p.Q252H, p.Y253H, p.E255K and p.E255V. Our tree reconstructions suggest that 1–3 ABL1-mutated clones may have occurred in cells already harbouring ABL1.p.Phe359Val, consistent with the notion that more than one ABL1 mutation can coexist within the same clone. Aggregated ABL1-mutated clones constituted 70–99% of cells at blast phase (Fig. 4e,f). Subclones showed several hundred extra genome-wide mutations, inconsistent with the short time that had passed (0.4 years), indicating earlier ABL1 mutation acquisition, rapid clonal outgrowth and/or genomic instability during blast phase.

Discussion

Most cancers follow a multi-step trajectory, with clones progressively acquiring genetic and epigenetic alterations that lead to a malignant tumour. This process commences decades, frequently 30–50 years, before cancer presentation^5,6,7. Against both these patterns, CML is an unusual neoplasm. Cancers driven by a single acquired genetic alteration are rare and we did not identify concurrent genomic alterations required for clonal expansion beyond BCR::ABL1. ‘One-hit’ cancers can be observed in the paediatric setting, with infantile MLL-rearranged acute lymphoblastic leukaemia³⁶ and paediatric ependymomas also driven by single genetic fusions^37,38. These tumours presumably occur during a susceptible developmental window. Beyond the paediatric setting, haematological cancers driven by single genetic events, such as mutant-JAK2 driven myeloproliferative neoplasms, are also characterized by slow clonal outgrowth over decades^7,9,32,39. The cancer trajectory to adult CML bucks this trend, with explosive and rapid tumour growth, reaching tumour doubling times of 2–3 weeks and clinical disease presenting as soon as 3–4 years later. In keeping with this, our data highlight an absence of BCR::ABL1 clonal haematopoiesis in the population. Subsequent genomic evolution following BCR::ABL1 drives disease progression, in line with previous studies^15,17,40, but also determines cell identity during blast phase.

Our estimates of the duration from the start of BCR::ABL1 clonal expansion to disease presentation fall within ranges from radiobiological studies of cancer incidences in Japanese survivors of the atomic bombs^8,41. This suggests that ionizing radiation increases CML incidence by means of increased initiation of BCR::ABL1, and that subsequent CML clonal trajectories disregard the initial mode of BCR::ABL1 acquisition. Indeed, such a short period from BCR::ABL1 acquisition to clinical presentation can only be explained by extremely rapid outgrowth. Our estimates of duration and rates of growth are inferred from independent parameters of the phylogenetic trees; specifically, the length of the shared branch harbouring BCR::ABL1 and the pattern of coalescent intervals during clonal expansion, respectively. We validate our estimates using orthogonal mathematical modelling and observe independent evidence of rapid growth through accelerated mutation acquisition and telomere attrition. Thus, the data presented offer robust inferences of CML clonal dynamics.

Within our small cohort, we observe variation in growth rates, with explosive growth (more than 10,000–1,000,000% per year) in three young patients and growth rates of 114 and 275% per year in two patients aged more than 60 years. Our cohort is too small to make definitive correlations but one may speculate as to possible reasons for this. Age could be a contributing factor through cell extrinsic factors, for example, competing clones of clonal haematopoiesis (PD51635) may hamper BCR::ABL1 clonal expansion in older patients. Cell intrinsic causes, such as shortened telomeres, age-related HSPC changes or atypical t(9;22) events (for example, loss of reciprocal ABL1::BCR, observed in PD51633 and PD51635)⁴² may also affect BCR::ABL1 potency. Variation in growth rates could affect disease response, as four out of the five individuals with the highest growth rates failed to achieve optimal therapeutic targets, in contrast to more slowly growing CML. The intricate link between the rate of decline of BCR::ABL1 levels post-TKI and the likelihood of achieving long-term remissions has long been recognized^43,44; however, the mechanistic basis underlying patient-to-patient variability in BCR::ABL1 kinetics has been lacking. Our data indicate that the rate of growth of the tumour may be an important factor. One in five patients with CML still fail two lines of TKI therapy⁴⁵ and early treatment failure is associated with a poor prognosis. Thus, a better understanding of the role of CML cancer trajectories could enable earlier personalization of therapy to enhance tumour remission.

Methods

Patients and sample acquisition

Peripheral blood and bone marrow samples were obtained from patients with chronic phase CML. We selected patients who responded to first-line TKI (PD51633, PD51632, PD51635 and PD57332), those that did not respond to first-line TKI (PD57334, PD57335 and PD57333) and those with slow response (PD56961 and PD51634). Patients with advanced-phase CML were also included. Patients provided informed, written consent for the use of their samples for research. Patients were selected to include a wide range of ages at diagnosis and variable treatment outcomes. The study was covered under NHS Research Ethics Committee approval numbers 05/MRE/44 and 18/EE/0199.

In vitro expansion of single-cell-derived blood colonies

MNCs were isolated from peripheral blood or bone marrow samples using Lymphoprep TM (Stem Cell Technologies). Single-cell suspensions of MNCs were grown in semisolid methylcellulose-based medium MethoCultTM H4034 Optimum (Stem Cell Technologies) for 10–14 days as previously described⁷. Colonies were individually picked and lysed in 45 μl of RLT buffer (Qiagen).

RT–PCR for BCR::ABL1 transcript

Transcript types from patients PD56961, PD57332 and PD57335 were determined using a standardized PCR with reverse transcription (RT–PCR) protocol⁴⁶. Primers used were BCR-e1-A: GACTGCAGCTCCAATGAGAAC (BCR exon 1), BCR-b1-A: GAAGTGTTTCAGAAGCTTCTCC (BCR exon 12–13) and ABL-a3-B: GTTTGGGCTTCACACCATTCC (ABL1 exon 3). This yielded a 344-bp PCR product for transcript e13a2 (6 bp (e12) + 106 bp (e13) + 175 bp (a2) + 57 bp (a3)) and a 419 bp PCR product for transcript e14a2 (6 bp (e12) + 106 (e13) + 75 (e14)+ 175 (a2) + 57 (a3)).

DNA library preparation, sequencing and read alignment

10–20 μl of lysed colony suspensions underwent WGS library preparation using the ‘laser capture micro-dissected biopsy’ pipeline⁴⁷ with eight cycles of PCR. This pipeline enables the generation of high complexity WGS libraries from an input of 150–200 cells. Samples with more than 2 ng μl⁻¹ of generated library DNA were used for paired-end, 150-bp reads WGS with Illumina NovaSeq 6000 machines. Reads were aligned to the human reference genome (GRCh38, NCBI) using the BWA-MEM (Burrows–Wheeler Aligner) algorithm.

Somatic mutation identification and filtering

SNV were identified using CaVEMan⁴⁸ for each colony by comparison to an in silico unmatched sample (PD38Is_wgs). CaVEMan was run with the ‘normal contamination of tumour’ parameter set to zero, and the tumour or normal copy numbers set to five or two. Reads supporting an SNV had to have a median BWA-MEM alignment score greater than or equal to 140 and median number of soft clipped bases of 0. Further filtering designed for this bespoke pipeline was applied (https://github.com/MathijsSanders/SangerLCMFiltering). The use of the unmatched normal meant that this process called both somatic and germline SNVs. The removal of germline SNVs and artefacts of sequencing required further filtering. As published in ref. ⁷, we used pooled information across colonies and read counts from a matched germline WGS buccal sample to ensure that genuine somatic variants that may be present in the germline sample, either as embryonic variants or due to tumour-in-normal contamination, were also identified. Short insertions and deletions (indels) were called using cgpPindel⁴⁹ with the standard WGS cgpPindel VCF filters applied, except the F018 Pindel filter was disabled as it excludes loci of depth of less than ten. Copy number aberrations (CNA) were identified using ASCAT⁵⁰ by comparison to a matched normal sample or a wild-type colony from the same individual. The union of colony SNVs and indels was then taken, and reads counted across all samples belonging to the individual (colonies and buccal samples) using VAFCorrect. This allows mutations detected in any one or more colonies to be identified across all other colonies to fully capture the pattern of sharing of mutations across the colonies from an individual patient. This generates a data matrix of the number of reads supporting every mutation, depth of the sequencing of that site, and the variant allele fraction across every colony for a patient. Brass and GRIDSS pipeline⁵¹ was used for calling structural variants. Both cgpPindel and Brass used PDv38is_wgs as an unmatched normal sample.

Creating a genotype matrix

The genotype at every locus within each sample was 1 (present), 0 (absent) or NA (unknown). We inferred the genotype in a depth sensitive manner. We assumed the observed mutant read count for a colony at a given site was MTR ~ binomial(n = depth, P = expected VAF), if the site was mutant, and MTR ~ binomial(n = depth, P = 0.01), if the site was wild-type. The genotype was set to the most likely of the two possible states provided one of the states was at least 20 times more likely than the other. Otherwise, the genotype was set to missing (NA). The expected VAF was 0.5 for autosomal sites, but for chromosome X, Y and CNA sites, it was set to 1/ploidy. For loss-of-heterozygosity sites, genotype was overridden and set to missing if it was originally 0.

Phylogenetic tree topology

We constructed phylogenetic tree topologies using maximum parsimony with MPBoot⁵². This method minimizes the number of changes required to reach the set of mutations assigned to each sample. The inputs for MPBoot were the binary genotype matrices with missing values per individual. These were exported as a multiple alignment fasta file with one line per colony with mutant represented as A, wild-type as T and missing as ?. The command line used was: mpboot -f <fasta>-bb 1,000.

Donor with no wild-type colonies

No wild-type colonies were captured for PD57332. The tree was constructed as above, which resulted in just the mutant clade being present. Approximate features of the mutant ‘truncal’ branch were inferred as follows: (1) estimating the duration of the mutant clade as the average root to tip SNV burden divided by the cohort average BCR::ABL1 SNV mutation rate, (2) the duration of the trunk was inferred (age at sample-height of the mutant clade) and (3) the length of the trunk in molecular time was then taken as the (duration of trunk) × (cohort wild-type SNV mutation rate) + (expected mutation burden at birth).

Driver annotation

As each branch on a phylogenetic tree was assigned with SNVs and indels, it was possible to screen all branches for the presence of potential driver mutations. For this purpose, a previously⁷ composed list of genes (n = 35), common in clonal haematopoiesis and myeloproliferative disorders, was used: ASXL1, BCOR, CALR, CBL, CSF3R, CUX1, DNMT3A, EZH2, GATA2, GNAS, GNB1, IDH1, IDH2, JAK2, KIT, KRAS, MLL3, MPL, NF1, NFE2, NRAS, PHF6, PPM1D, PTPN11, RB1, RUNX1, SETBP1, SF3B1, SRSF2, SH2B3, STAG2, TET2, TP53, U2AF1 and ZRSR2, with the addition of ABL1 and BCR, totalling 37 genes. Branches with identified drivers were highlighted in colour on the phylogenetic trees. Annotation for BCR::ABL1 fusion was added manually to the branch on the basis of its presence or absence in individual colonies from Brass and GRIDSS results.

Timing branches

Given the linear accumulation of somatic mutations with age, we can infer the time point in life when driver mutations in phylogenetic trees had occurred. Branches at the top of a tree comprise mutations acquired at a young age, with branches lower down representing mutations arising later in life. We used our previously developed method rtreefit (https://github.com/nangalialab/rtreefit) for converting trees in which branch lengths are expressed in molecular time (that is, number of mutations) into trees in which the branch lengths are expressed in units of time (years)⁷. In brief, the method jointly fits separate wild-type and mutant constant mutation rates (that is, number of SNVs accumulated per year) and absolute time branch lengths using a Bayesian per individual tree-based model under the assumption that the number of observed mutations assigned to a branch is Poisson distributed with mean = branch duration × sensitivity × mutation rate, and subject to the constraint that the root to tip duration is equal to the age at sampling. Furthermore, the method accounts for an elevated mutation rate during embryogenesis by assuming an excess mutation rate through development. In running rtreefit, the mutant clade was defined as not including the trunk, so the method assumed that BCR-ABL1 was acquired at the end of the trunk. The donor PD57332 had no wild-type samples so an in silico wild-type outgroup with a long branch (equal to 1,000 × cohort wild-type mutation rate) was added. This enabled PD57332 to be processed by rtreefit in a similar way to the other donors. The rtreefit algorithm was run with four chains and 20,000 iterations per chain. Mutations were assigned to the tree in a depth sensitive manner using treemut with mutations being hard-assigned to the highest probability branch (https://github.com/nangalialab/treemut). Branch lengths were adjusted for the branch-specific SNV detection sensitivity⁷, in which the sensitivity of detection of fully clonal SNV variants was directly estimated from the per colony sensitivity for detecting germline heterozygous SNVs together with a multiplicative correction for the clonality (VAF) of the colonies. In calculating mutation burden and branch lengths, CNAs present in any colony in an individual were uniformly masked in all colonies for that individual and then the overall mutation burden was scaled back up by the reciprocal of 1 − expected number of mutations in the masked region. In addition to SNVs, indels and the BCR-ABL1 fusion, colonies showed a variety of CNA events. These events were curated as being present or absent in each of the colonies giving an event genotype vector like that obtained for SNVs and indels. Once the tree topology was inferred using the SNV genotypes, the branches that exactly matched the event genotype were identified and the event assigned to the corresponding branch.

Quality control of a phylogenetic tree topology

The initial quality assessment step of phylogenetic trees included the removal of colonies for which the sensitivity of CaVEMan somatic mutation detection was below 60%. Colonies that might have been contaminated with cells from another colony were also excluded. If the colony was not clonal, then the mean VAF of SNVs assigned to its private branch would be lower than 50%. To ensure that samples were clonal, colonies were excluded if the VAF of SNVs that mapped to their private branches was significantly lower than 0.4. If the VAF of mutations assigned to a colony was not 0 in non-ancestral branches, then that colony was also removed because of this indicating that cells from this colony were contaminating other colonies.

Per sample BCR::ABL1 mutation status

To ascertain BCR::ABL1 mutation status for each sample, we undertook a ‘joint’ genotyping methodology using GRIDSS. For each patient, Brass structural variant calls were reviewed for all events of interest pertaining to the BCR::ABL1 fusion event that included a core ‘target’ region consisting of the gene regions of ABL1 and BCR. For PD51635, we added RBFOX3 (chromosome 17), and for PD51633, we incorporated the large deletions present on der(9). For each patient, all samples were analysed together over the target regions to jointly identify evidence for structural variants. Per-patient results were reviewed and visualized with gGnome (https://github.com/mskilab-org/gGnome), which allowed the reconstruction of derivative chromosomes in patients with complex events such as PD51635. To ascertain the consequence of translocations, GRASS (Gene Rearrangement Analysis System, https://github.com/cancerit/grass) was used. In PD51635, a complex structural variant t(9;17) event (inversion ‘chr. 9 130776970GG]chr9:130786796]’ and translocation ‘chr. 9 130777016T [chr17:79184121[ATT’) was simplified to an equivalent single translocation event (chr. 9: +:130777016, chr. 17: +:79184121, AT) for annotation (Fig. 1d and Supplementary Table 2).

Prediction of consequence of exonic BCR::ABL1 breakpoints

In three patients (PD56961, PD57332 and PD57335) we identified breakpoints in BCR exons (exon 14, 15 and 15, respectively). To identify stop codons, we continued the reading frame from the breakpoint, adjusting for any inserted bases within the translocation event. We used SpliceAI⁵³ to predict the splicing probabilities for each respective fusion sequence. Fusion sequences were reconstructed using GRCh38 reference sequence for BCR (chr. 22: 23170509–23328037) and ABL1 (chr. 9: 130825254–130897675) incorporating a 10-kb flanking sequence and any extra bases from the BCR::ABL1 translocation event (Extended Data Fig. 2a). Reconstructed fusion and reference sequences were used as input in the ‘custom sequence’ script (https://github.com/Illumina/SpliceAI), and ‘raw’ splice acceptor and donor probabilities from SpliceAI were converted to bedGraph format and reviewed on IGV⁵⁴.

Mutational signature analysis

De novo mutation signature extraction was performed using HDP (https://github.com/nicolaroberts/hdp) with the mutations assigned to individual branches being treated as samples. Branches with fewer than 50 mutations were grouped into two per donor groups; those short branches that end before 100 mutations molecular time were pooled in the ‘early life’ group and the rest of the short branches were into the ‘late life’ group. These groups were also treated as samples. HDP extraction was then run across four chains with the following parameters for hdp_posterior: burnin=10,000, n=500, spacing=250. SBSblood signature⁵⁵ was downloaded and collated with the Pan-Cancer Analysis of Whole Genomes signatures (https://cog.sanger.ac.uk/cosmic-signatures-production/documents/COSMIC_v3.3.1_SBS_GRCh38.txt). De novo extraction identified three signatures SBSblood, SBS1 and SBS18, which showed the following cosine similarities to their respective published Pan-Cancer Analysis of Whole Genomes and/or SBSblood signatures: 0.927, 0.942 and 0.884. We refitted per-donor branch groupings and individual branches against the above published versions of SBSblood, SBS1 and SBS18. This signature attribution was carried out for each of these per-donor categories or branches using sigfit::fit_to_signature with the default ‘multinomial’ model. The per-branch attributions were then carried out by (1) assigning a per-mutation signature membership probability and then (2) summing these signature membership probabilities over all SNVs assigned to a branch to obtain a branch-level signature attribution proportion. The per-mutation signature membership probability was calculated using:

$$P({\rm{m}}{\rm{u}}{\rm{t}}{\rm{a}}{\rm{t}}{\rm{i}}{\rm{o}}{\rm{n}}\,{\epsilon }\,{\rm{S}}{\rm{i}}{\rm{g}})=\frac{P({\rm{m}}{\rm{u}}{\rm{t}}{\rm{a}}{\rm{t}}{\rm{i}}{\rm{o}}{\rm{n}}\,|{\rm{S}}{\rm{i}}{\rm{g}}){P}_{0}({\rm{S}}{\rm{i}}{\rm{g}})}{{\sum }_{{{\rm{S}}{\rm{i}}{\rm{g}}}^{{\prime} }{\epsilon }\{{\rm{S}}{\rm{B}}{\rm{S}}1,{\rm{S}}{\rm{B}}{\rm{S}}{\rm{b}}{\rm{l}}{\rm{o}}{\rm{o}}{\rm{d}},{\rm{S}}{\rm{B}}{\rm{S}}18\}}P({\rm{m}}{\rm{u}}{\rm{t}}{\rm{a}}{\rm{t}}{\rm{i}}{\rm{o}}{\rm{n}}\,|{{\rm{S}}{\rm{i}}{\rm{g}}}^{{\prime} }){P}_{0}({{\rm{S}}{\rm{i}}{\rm{g}}}^{{\prime} })}$$

where the prior probability, P₀(Sig), is given by the mean Sigfit attribution probability of the specified signature, Sig, for the category that the mutation belongs to. The cohort level analysis of mutational signatures and C>T at CpG representation for branch categories was carried out using a random effects meta-analysis using the rma function in the ‘metafor’ R package.

Growth rate estimation

The growth rate of BCR-ABL1 clones was estimated using the previously described phylofit approach^7,9. In brief, phylofit is a Bayesian approach that estimates the growth rate by directly fitting a three-parameter logistic growth curve trajectory using the joint probability density of coalescence times given the population size trajectory. The three parameters estimated by this method are the saturation population size N, the exponential phase growth rate s and the midpoint of the curve t^(m). Given that patients with CML present with a high BCR::ABL1 fraction we set the upper bound of the prior for the midpoint to be the age at diagnosis and the lower bound to be the age of the MRCA of the mutant clade. We adopted uniform priors on the growth rate, s, and on the log scale saturation population size. Now, the annualized growth rate is given by S = exp(s) − 1 and the uniform priors for s, midpoint t^(m) and population size N, are:

$$\begin{array}{c}\,\,\,\,s \sim {\rm{U}}{\rm{n}}{\rm{i}}{\rm{f}}{\rm{o}}{\rm{r}}{\rm{m}}(0.001,30)\\ \,\,\,{t}^{(m)} \sim {\rm{U}}{\rm{n}}{\rm{i}}{\rm{f}}{\rm{o}}{\rm{r}}{\rm{m}}({\rm{a}}{\rm{g}}{\rm{e}}\,{\rm{o}}{\rm{f}}\,{\rm{M}}{\rm{R}}{\rm{C}}{\rm{A}},{\rm{a}}{\rm{g}}{\rm{e}}\,{\rm{a}}{\rm{t}}\,{\rm{d}}{\rm{i}}{\rm{a}}{\rm{g}}{\rm{n}}{\rm{o}}{\rm{s}}{\rm{i}}{\rm{s}})\\ \,{\log }_{10}(N) \sim {\rm{U}}{\rm{n}}{\rm{i}}{\rm{f}}{\rm{o}}{\rm{r}}{\rm{m}}(4,7)\end{array}$$

The choice of upper bound for t^(m) was motivated by the observation that patients with CML generally show a high BCR-ABL1 burden at diagnosis. The input data for phylofit was the time-based trees obtained using rtreefit as described above. Note that the branch lengths of the input trees were chosen to be the mean of the posterior branch lengths. The trees were restricted to the earliest sampling point. These time points are all at diagnosis or after diagnosis. Comparison of estimates were checked using an alternative recently published method cloneRate²⁴ as detailed in Supplementary Note 3.

Telomere analysis

Telomerecat (v.4.0.2, https://github.com/cancerit/telomerecat) was used to estimate mean telomere length (bp) with (-t 75) to ameliorate the impact of NovaSeq sequencing artefacts (further details in Supplementary Note 2).

Mixed models

Linear mixed models used for SNV burden and telomere analysis were implemented in the R package lme4 to estimate the impact of age and mutant status. Age (age_at_sample_exact) was defined as the count of completed years from birth at sampling and sample mutant status (BCR_ABL1) was defined as BCR::ABL1 positive (Mt) or negative (Wt). For each response variable (y), we first tested the significance of mutant status (model 0 versus model 1) and then further terms were added to see whether this improved the model compared to the base model (model 1).

$$\begin{array}{l}{\rm{model}}\_0=y \sim 1+{\rm{age}}\_{\rm{at}}\_{\rm{sample}}\_{\rm{exact}}+(1| {\rm{patient}})\\ {\rm{model}}\_1=y \sim 1+{\rm{age}}\_{\rm{at}}\_{\rm{sample}}\_{\rm{exact}}+BCR\_ABL1+(1| {\rm{patient}})\\ {\rm{model}}\_2=y \sim 1+{\rm{age}}\_{\rm{at}}\_{\rm{sample}}\_{\rm{exact}}+BCR\_ABL1\\ \,\,\,\,\,+(1+{\rm{age}}\_{\rm{at}}\_{\rm{sample}}\_{\rm{exact}}| {\rm{patient}})\\ {\rm{model}}\_3=y \sim 1+{\rm{age}}\_{\rm{at}}\_{\rm{sample}}\_{\rm{exact}}+BCR\_{\rm{ABL}}1\\ \,\,\,\,\,+(1+BCR\_ABL1| {\rm{patient}})\end{array}$$

Models were fitted with default lme4 parameters. If a model did not converge, lme4::allFit() was used to refit the model to all available optimizers (provide by the lme4, optimx and dfoptim R packages), the best optimizer was selected from non-singular and converged refits with the highest negative log-likelihood. Only non-singular and converged models were considered for model selection using the Bayesian information criterion (BIC). For the final selected model, 95% confidence intervals (percentile bootstrap intervals) were calculated for each fixed effect, using confint(type=‘perc’) from the first 1,000 converged and non-singular parametric bootstrapped models generated using the bootstrap() function from the R package lmeresampler, using a seed (1234) for reproducibility.

SNV burden models

We first removed all samples (n = 35) from PD57332 as we were only able to grow BCR::ABL1 colonies and therefore our estimations of SNV mutation burden (nsub_adj) were expected to be biased; this left 799 samples across eight patients. The final model used is shown below and was found to have the lowest BIC value (8,645.52), as detailed in Supplementary Note 1.

$$\begin{array}{l}{\rm{SNV}}\;{\rm{model}}\,3={\rm{nsub}}\_{\rm{adj}} \sim {\rm{age}}\_{\rm{at}}\_{\rm{sample}}\_{\rm{exact}}+BCR\_ABL1\\ \,\,\,\,\,\,+(1+BCR\_ABL1| {\rm{patient}})\end{array}$$

Telomere length models

To account for reported issues with telomere estimation using Telomerecat and NovaSeq sequenced data⁹, we explored any batch effect of library preparation (library.cluster) and sequencing run (run_id.uniq) within wild-type samples (n = 469) on mean telomere length (length). The addition of sequencing run as an extra random effect (1| run_id.uniq) resulted in the lowest BIC model (7,483.33). This term was added to a series of models tested on the full dataset (n = 834), with the final model below having the lowest BIC value (13,265.15), as detailed further in Supplementary Note 2.

$$\begin{array}{l}{\rm{Telomere}}\;{\rm{model}}\,3={\rm{length}} \sim {\rm{age}}\_{\rm{at}}\_{\rm{sample}}\_{\rm{exact}}+BCR\_ABL1\\ +(1+BCR\_ABL1| {\rm{patient}})+(1| {\rm{run}}\_{\rm{id}}.\,{\rm{uniq}})\end{array}$$

Bulk DNA library preparation, sequencing and read alignment

DNA extracted from peripheral blood was subjected to WGS library preparation and sequenced paired-end, with 150-bp reads on Illumina NovaSeq 6000 machines. The reads were aligned to the human reference genome (GRCh38, NCBI) using the BWA-MEM algorithm.

Bulk unmatched somatic mutation identification and filtering

SNVs were identified using CaVEMan⁴⁷ for each bulk sample by comparison to an in silico unmatched sample (PD38Is_wgs). CaVEMan was run with the ‘normal contamination of tumour’ parameter set to zero, and the tumour and normal copy numbers set to five and two, respectively. To increase sensitivity, SNVs only flagged as seen in a panel of normal samples (‘VUM’) were rescued. All SNVs were required to have less than half of supporting reads clipped (CLPM = 0) and a median BWA-MEM alignment score greater than or equal to 140 (ASMD ≥ 140). Short insertions and deletions (indels) were called using cgpPindel⁴⁸ with standard WGS cgpPindel VCF filters applied, except the F010 Pindel filter as it excludes variants seen in a panel of normal samples. Driver candidate variants were restricted to the 37 genes described above. To filter germline variants, we retained SNVs and indels with a gnomAD v.3.1.2 (ref. ⁵⁶) popmax allele frequency less than 0.01 (annotated using echtvar v.0.2.0, ref. ⁵⁷). The union of SNVs and indels was taken and reads counted across all samples belonging to the individual using VAFCorrect as above. Sequencing read depth was reviewed in IGV⁵⁴ to identify arm-level copy number events. GRIDSS v.2.13.1 was used for structural variant calling, and to estimate the VAF of BCR::ABL1.

Bulk phylogeny reconstruction of PD60243

We used DPClust^58,59 to infer mutational clusters using SNVs (CaVEMan) and copy number and/or sample purity (Battenberg) called for each sample of PD60243a (81.3 years granulocytes, PD60243c 81.7 years mononuclear cells, PD60243d 81.7 years granulocytes) using a matched buccal sample (PD60243e). Manual inspection of the initial clusters identified 19 outlier poor quality variants that were subsequently removed for a second DPClust run with the addition of all ABL1 mutations identified in the unmatched analysis. All four mutually exclusive ABL1 mutations (as determined by read phasing) were grouped in the same cluster meaning we could not discern the exact subclonal phylogeny of ABL1 mutations (which is expected with bulk reconstruction given the low VAF and 80× coverage). We inferred the total ABL1 mutation burden at blast phase by adding the mutually exclusive ABL1 mutations as individual subclones and removing their assigned cluster. Using ctree (https://github.com/caravagnalab/ctree)⁶⁰, we performed an exhaustive tree search (sspace.cutoff of 100,000) and retained the top scoring trees that maintained ABL1 mutation mutual exclusivity. We calculated the sum of ABL1 cancer cell fractions (CCF), accounting for nesting, to obtain the CCF median and range.

All of Us cohort analysis

The All of Us Research Program²⁹ is a US population-based cohort that links electronic health record data to WGS data for 206,173 participants, with an average genome-wide coverage of 30×. We searched CRAM files (Data repository v.7) to identify sequencing reads supporting a canonical BCR::ABL1 (or reciprocal ABL1::BCR) variant between BCR (chr. 22: 23289313–23292813) and ABL1 (chr. 9: 130674613–130874613). Reads that were incorrectly oriented or had insufficient mapping quality (score less than or equal to 6) were filtered. GRASS (https://github.com/cancerit/grass) was used to annotate sequencing reads that passed filters. GRASS takes pairs of genomic coordinates representing potential rearrangement events and predicts the fusion consequences along with their associated genes. We required the fusion event to be specifically between BCR and ABL1. After identifying BCR::ABL1 carriers in the cohort, we extracted electronic health record data and searched for International Classification of Diseases (ICD) codes related to cancer, blood or immune diseases. In addition to searching for ‘chronic myeloid leukaemia’ (CML), we included conditions such as ‘abnormal white cell count’, ‘basophilia’, ‘splenomegaly’, ‘unspecified cancer’ or ‘anaemia’. On the basis of these disease entries and domain knowledge, we categorized carriers into four groups: (1) CML mentioned, (2) abnormal white count (for example, leucocytosis), basophilia or splenomegaly mentioned, (3) other haematological issues (for example, anaemia) or unspecified cancer and (4) no relevant disease. Blood sampling date and the date of the most recent ICD code was used to define the time from sample collection to last follow-up. When calculating the incidence of CML and BCR::ABL1 carrier status across age groups, we defined age at diagnosis for CML and age at blood sampling for BCR::ABL1 carrier status. For each age group, summary statistics were calculated by averaging annual statistics from 2018 to 2022, assuming population structure remained stable during this period.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Sequencing files are available through the EGA (Dataset EGAD00001015353) in line with Wellcome Sanger Institute data sharing, and all somatic mutation .vcf files have been uploaded to Mendeley (https://doi.org/10.17632/yg29vx2f35.1). Use of individual-level data in the All of Us Program is available to researchers across the world through the Researcher Workbench, a cloud-based computing platform (https://www.researchallofus.org/register/). Summary-level data are available to the public through a data browser provided by the research programme (https://databrowser.researchallofus.org/). Source data are provided with this paper.

Code availability

Code and software created for analyses are available at GitHub (https://github.com/nangalialab/CML).

References

Nowell, P. C. & Hungerford, D. A. A minute chromosome in human chronic granulocytic leukemia. Science 132, 1497 (1960).
Rowley, J. D. Anew consistent chromosomal abnormality in chronic myelogenous leukaemia identified by quinacrine fluorescence and giemsa staining. Nature 243, 290–293 (1973).
Article ADS CAS PubMed Google Scholar
Bower, H. et al. Life expectancy of patients with chronic myeloid leukemia approaches the life expectancy of the general population. J. Clin. Oncol. 34, 2851–2857 (2016).
Article CAS PubMed Google Scholar
Nowell, P. C. The clonal evolution of tumor cell populations. Science 194, 23–28 (1976).
Article ADS CAS PubMed Google Scholar
Mitchell, T. J. et al. Timing the landmark events in the evolution of clear cell renal cell cancer: TRACERx renal. Cell 173, 611–623.e17 (2018).
Article CAS PubMed PubMed Central Google Scholar
Gerstung, M. et al. The evolutionary history of 2,658 cancers. Nature 578, 122–128 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Williams, N. et al. Life histories of myeloproliferative neoplasms inferred from phylogenies. Nature 602, 162–168 (2022).
Article ADS CAS PubMed Google Scholar
Radivoyevitch, T., Hlatky, L., Landaw, J. & Sachs, R. K. Quantitative modeling of chronic myeloid leukemia: Insights from radiobiology. Blood 119, 4363–4371 (2012).
Article CAS PubMed PubMed Central Google Scholar
Mitchell, E. et al. Clonal dynamics of haematopoiesis across the human lifespan. Nature 606, 343–350 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Lee-Six, H. et al. Population dynamics of normal human blood inferred from somatic mutations. Nature 561, 473–478 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Dash, D. P., Trap-stamborski, V., Czajkowski, C., Rauscher, D. & Pillai, R. Identification of a novel BCR-ABL1 fusion transcript in a chronic myeloid leukemia (CML) patient by unique molecular methods and subsequent monitoring of the patient’s response to tyrosine kinase inhibitor (TKI) therapy. Blood 132, 5453 (2018).
Article Google Scholar
van der Velden, V. H. J., Beverloo, H. B., Hoogeveen, P. G. & Zwaan, C. M. A novel BCR-ABL fusion transcript (e18a2) in a child with chronic myeloid leukemia. Leukemia 21, 833–835 (2007).
Article PubMed Google Scholar
Bidikian, A. et al. Prognostic impact of ASXL1 mutations in chronic phase chronic myeloid leukemia. Blood Cancer J. 12, 144 (2022).
Article PubMed PubMed Central Google Scholar
Schönfeld, L. et al. ASXL1 mutations predict inferior molecular response to nilotinib treatment in chronic myeloid leukemia. Leukemia 36, 2242–2249 (2022).
Article PubMed PubMed Central Google Scholar
Branford, S. et al. Integrative genomic analysis reveals cancer-associated mutations at diagnosis of CML in patients with high-risk disease. Blood 132, 948–961 (2018).
Article CAS PubMed Google Scholar
Shanmuganathan, N. et al. Impact of additional genetic abnormalities at diagnosis of chronic myeloid leukemia for first-line imatinib-treated patients receiving proactive treatment intervention. Haematologica 108, 2380–2395 (2023).
CAS PubMed PubMed Central Google Scholar
Branford, S. et al. Laying the foundation for genomically-based risk assessment in chronic myeloid leukemia. Leukemia 33, 1835–1850 (2019).
Article PubMed PubMed Central Google Scholar
Osorio, F. G. et al. Somatic mutations reveal lineage relationships and age-related mutagenesis in human hematopoiesis. Cell Rep. 25, 2308–2316.e4 (2018).
Article CAS PubMed PubMed Central Google Scholar
Spencer Chapman, M. et al. Lineage tracing of human development through somatic mutations. Nature 595, 85–90 (2021).
Article ADS CAS PubMed Google Scholar
Slupianek, A. et al. BCR-ABL1 kinase inhibits uracil DNA glycosylase UNG2 to enhance oxidative DNA damage and stimulate genomic instability. Leukemia 27, 629–634 (2013).
Article CAS PubMed Google Scholar
Alexandrov, L. B. et al. Clock-like mutational processes in human somatic cells Europe PMC Funders Group. Nat. Genet. 47, 1402–1407 (2015).
Article CAS PubMed PubMed Central Google Scholar
Coorens, T. H. H. et al. Inherent mosaicism and extensive mutation of human placentas. Nature 592, 80–85 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Williams, N. nangalialab/rtreefit. Fit mutation rates and ultrametric trees to trees with known observed mutation count and calling sensitivity. GitHub https://github.com/nangalialab/rtreefit (2024).
Johnson, B., Shuai, Y., Schweinsberg, J. & Curtius, K. cloneRate: fast estimation of single-cell clonal dynamics using coalescent theory. Bioinformatics 39, btad561 (2023).
Article CAS PubMed PubMed Central Google Scholar
Clarke, C. J. & Holyoake, T. L. Preclinical approaches in chronic myeloid leukemia: from cells to systems. Exp. Hematol. 47, 13–23 (2017).
Article PubMed PubMed Central Google Scholar
Bernstein, N. et al. Analysis of somatic mutations in whole blood from 200,618 individuals identifies pervasive positive selection and novel drivers of clonal hematopoiesis. Nat. Genet. 56, 1147–1155 (2024).
Article CAS PubMed PubMed Central Google Scholar
Stacey, S. N. et al. Genetics and epidemiology of mutational barcode-defined clonal hematopoiesis. Nat. Genet. https://doi.org/10.1038/s41588-023-01555-z (2023).
Biernaux, C., Loos, M., Sels, A., Huez, G. & Stryckmans, P. Detection of major BCR-ABL gene expression at a very low level in blood cells of some healthy individuals. Blood 86, 3118–3122 (1995).
Article CAS PubMed Google Scholar
Bick, A. G. et al. Genomic data in the All of Us Research Program. Nature 627, 340–346 (2024).
Article Google Scholar
National Cancer Institute. Chronic myeloid leukemia—cancer stat facts. SEER https://seer.cancer.gov/statfacts/html/cmyl.html (2025).
Machado, H. E. et al. Diverse mutational landscapes in human lymphocytes. Nature 608, 724–732 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Fabre, M. A. et al. The longitudinal dynamics and natural history of clonal haematopoiesis. Nature 606, 335–342 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Hehlmann, R. et al. High-risk additional chromosomal abnormalities at low blast counts herald death by CML. Leukemia 34, 2074–2086 (2020).
Article CAS PubMed PubMed Central Google Scholar
Brioli, A. et al. Management and outcome of patients with chronic myeloid leukemia in blast phase in the tyrosine kinase inhibitor era—analysis of the European LeukemiaNet Blast Phase Registry. Leukemia 38, 1072–1080 (2024).
Article PubMed PubMed Central Google Scholar
Ochi, Y. et al. Clonal evolution and clinical implications of genetic abnormalities in blastic transformation of chronic myeloid leukaemia. Nat. Commun. 12, 2833 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Andersson, A. K. et al. The landscape of somatic mutations in infant MLL-rearranged acute lymphoblastic leukemias. Nat. Genet. 47, 330–337 (2015).
Article CAS PubMed PubMed Central Google Scholar
Parker, M. et al. C11orf95-RELA fusions drive oncogenic NF-κB signalling in ependymoma. Nature 506, 451–455 (2014).
Article ADS CAS PubMed PubMed Central Google Scholar
Pajtler, K. W. et al. YAP1 subgroup supratentorial ependymoma requires TEAD and nuclear factor I-mediated transcriptional programmes for tumorigenesis. Nat. Commun. 10, 3914 (2019).
Article ADS PubMed PubMed Central Google Scholar
Van Egeren, D. et al. Reconstructing the lineage histories and differentiation trajectories of individual cancer cells in myeloproliferative neoplasms. Cell Stem Cell 28, 514–523.e9 (2021).
Article PubMed PubMed Central Google Scholar
Ochi, Y. Genetic landscape of chronic myeloid leukemia. Int. J. Hematol. 117, 30–36 (2023).
Article CAS PubMed Google Scholar
Hsu, W.-L. et al. The incidence of leukemia, lymphoma and multiple myeloma among atomic bomb survivors: 1950–2001. Radiat. Res. 179, 361–382 (2013).
Article ADS CAS PubMed Google Scholar
Huntly, B. J. et al. Deletions of the derivative chromosome 9 occur at the time of the Philadelphia translocation and provide a powerful and independent prognostic indicator in chronic myeloid leukemia. Blood 98, 1732–1738 (2001).
Article CAS PubMed Google Scholar
Branford, S. et al. Prognosis for patients with CML and >10% BCR-ABL1 after 3 months of imatinib depends on the rate of BCR-ABL1 decline. Blood 124, 511–518 (2014).
Article CAS PubMed Google Scholar
Shanmuganathan, N. et al. Early BCR-ABL1 kinetics are predictive of subsequent achievement of treatment-free remission in chronic myeloid leukemia. Blood 137, 1196–1207 (2021).
Article CAS PubMed Google Scholar
Benmerad, M. et al. Tyrosine kinase inhibitor (TKI) treatment patterns in 5,261 chronic phase (CP) chronic myeloid leukemia (CML) patients in ‘real-life’ settings: a population-based study using a nationwide claim database. Blood 142, 3174 (2023).
Article Google Scholar
Van Dongen, J. et al. Standardized RT-PCR analysis of fusion gene transcripts from chromosome aberrations in acute leukemia for detection of minimal residual disease: report of the BIOMED-1 Concerted Action: investigation of minimal residual disease in acute leukemia. Leukemia 13, 1901–1928 (1999).
Article PubMed Google Scholar
Ellis, P. et al. Reliable detection of somatic mutations in solid tissues by laser-capture microdissection and low-input DNA sequencing. Nat. Protoc. 16, 841–871 (2021).
Article CAS PubMed Google Scholar
Jones, D. et al. cgpCaVEManWrapper: simple execution of CaVEMan in order to detect somatic single nucleotide variants in NGS data. Curr. Protoc. Bioinform. 56, 15.10.1–15.10.18 (2016).
Article Google Scholar
Ye, K., Schulz, M. H., Long, Q., Apweiler, R. & Ning, Z. Pindel: a pattern growth approach to detect break points of large deletions and medium sized insertions from paired-end short reads. Bioinformatics 25, 2865–2871 (2009).
Article CAS PubMed PubMed Central Google Scholar
Van Loo, P. et al. Allele-specific copy number analysis of tumors. Proc. Natl Acad. Sci. USA 107, 16910–16915 (2010).
Article ADS PubMed PubMed Central Google Scholar
Cameron, D. L. et al. GRIDSS: sensitive and specific genomic rearrangement detection using positional de Bruijn graph assembly. Genome Res. 27, 2050–2060 (2017).
Article CAS PubMed PubMed Central Google Scholar
Hoang, D. T. et al. MPBoot: fast phylogenetic maximum parsimony tree inference and bootstrap approximation. BMC Evol. Biol. 18, 11 (2018).
Article PubMed PubMed Central Google Scholar
Jaganathan, K. et al. Predicting splicing from primary sequence with deep learning. Cell 176, 535–548.e24 (2019).
Article CAS PubMed Google Scholar
Robinson, J. T. et al. Integrative genomics viewer. Nat. Biotechnol. 29, 24–26 (2011).
Article CAS PubMed PubMed Central Google Scholar
Machado, H. E. et al. Convergent somatic evolution commences in utero in a germline ribosomopathy. Nat. Commun. 14, 5092 (2023).
Article ADS CAS PubMed PubMed Central Google Scholar
Chen, S. et al. A genomic mutational constraint map using variation in 76,156 human genomes. Nature 625, 92–100 (2024).
Article ADS CAS PubMed Google Scholar
Pedersen, B. S. & de Ridder, J. Echtvar: compressed variant representation for rapid annotation and filtering of SNPs and indels. Nucleic Acids Res. 51, e3 (2023).
Article CAS PubMed Google Scholar
Nik-Zainal, S. et al. The life history of 21 breast cancers. Cell 149, 994–1007 (2012).
Article CAS PubMed PubMed Central Google Scholar
Dentro, S. C., Wedge, D. C. & Van Loo, P. Principles of reconstructing the subclonal architecture of cancers. Cold Spring Harb. Perspect. Med. 7, a026625 (2017).
Article PubMed PubMed Central Google Scholar
Caravagna, G. et al. Detecting repeated cancer evolution from multi-region tumor sequencing data. Nat. Methods 15, 707–714 (2018).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank the Cambridge Stem Cell Biobank for their invaluable support with sample collection and colony growth. We acknowledge the contribution of the Cancer, Ageing and Somatic Mutation laboratory and IT teams, and the Sanger core facility for their support with next generation sequencing. We thank J. Jones, H. Langoallen, the haemato-oncology diagnostics service and the East Genomics Laboratory Hub, Cambridge University Hospitals for diagnostic services. The Nangalia laboratory is supported by Cancer Research UK, Wellcome core funding at Wellcome Sanger Institute, Alborada Trust, the Leukaemia Lymphoma Society and Blood Cancer UK. We thank the patients for their participation in this study. We gratefully acknowledge All of Us participants for their contributions, without whom this research would not have been possible. We thank the National Institutes of Health’s All of Us Research Program for making available the participant data examined in this study.

Author information

These authors contributed equally: Aleksandra E. Kamizela, Daniel Leongamornlert

Authors and Affiliations

Wellcome Sanger Institute, Hinxton, UK
Aleksandra E. Kamizela, Daniel Leongamornlert, Nicholas Williams, Xin Wang, Kudzai Nyamondo, Kevin Dawson, Michael Spencer Chapman, Jing Guo, Joe Lee, Karim Mane, Peter J. Campbell & Jyoti Nangalia
Cambridge Stem Cell Institute and Department of Haematology, University of Cambridge, Cambridge, UK
Aleksandra E. Kamizela, Kudzai Nyamondo, Joe Lee, Karim Mane, Anthony R. Green, Peter J. Campbell, Brian J. P. Huntly & Jyoti Nangalia
Cambridge University Hospitals NHS Trust, Cambridge, UK
Aleksandra E. Kamizela, Kudzai Nyamondo, Kate Milne, Anthony R. Green, Brian J. P. Huntly, E. Joanna Baxter & Jyoti Nangalia
Cardiovascular Disease Initiative, Broad Institute of MIT and Harvard, Cambridge, MA, USA
Xin Wang & Patrick T. Ellinor
Centre for Haemato-Oncology, Barts Cancer Institute, Queen Mary University of London, London, UK
Michael Spencer Chapman
Department of Clinical and Experimental Medicine, Brighton and Sussex Medical School, Brighton, UK
Timothy Chevassut
Cardiovascular Research Center, Massachusetts General Hospital, Boston, MA, USA
Patrick T. Ellinor

Authors

Aleksandra E. Kamizela
View author publications
Search author on:PubMed Google Scholar
Daniel Leongamornlert
View author publications
Search author on:PubMed Google Scholar
Nicholas Williams
View author publications
Search author on:PubMed Google Scholar
Xin Wang
View author publications
Search author on:PubMed Google Scholar
Kudzai Nyamondo
View author publications
Search author on:PubMed Google Scholar
Kevin Dawson
View author publications
Search author on:PubMed Google Scholar
Michael Spencer Chapman
View author publications
Search author on:PubMed Google Scholar
Jing Guo
View author publications
Search author on:PubMed Google Scholar
Joe Lee
View author publications
Search author on:PubMed Google Scholar
Karim Mane
View author publications
Search author on:PubMed Google Scholar
Kate Milne
View author publications
Search author on:PubMed Google Scholar
Anthony R. Green
View author publications
Search author on:PubMed Google Scholar
Timothy Chevassut
View author publications
Search author on:PubMed Google Scholar
Peter J. Campbell
View author publications
Search author on:PubMed Google Scholar
Patrick T. Ellinor
View author publications
Search author on:PubMed Google Scholar
Brian J. P. Huntly
View author publications
Search author on:PubMed Google Scholar
E. Joanna Baxter
View author publications
Search author on:PubMed Google Scholar
Jyoti Nangalia
View author publications
Search author on:PubMed Google Scholar

Contributions

The following authors contributed equally: A.E.K. and D.L.; N.W. and X.W. A.E.K., N.W. and D.L. performed all genomic analyses. K.D. assisted with modelling. J.L., K.N. and K. Milne assisted with patient selection, and sequencing. E.J.B. directed colony growth. K. Mane provided computational support. T.C. provided mentorship to A.E.K. A.R.G., P.J.C. and B.J.P.H. provided scientific advice. X.W., J.G. and P.T.E. analysed All of Us data. M.S.C. performed BCR:ABL1 screening in colonies. A.E.K., N.W., D.L., X.W. and J.N. prepared the manuscript. J.N. conceived the study and directed the research.

Corresponding author

Correspondence to Jyoti Nangalia.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature thanks Jane Apperley and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 Patient cohort details with BCR::ABL1/ABL1 ratios during treatment.

Plots showing the clinical BCR::ABL1/ABL1 ratio (international standard levels) measured using RT-qPCR from diagnosis until last follow-up. The colours on the plot indicate an optimal (green), warning (yellow) or failure (red) of response according to European Leukemia Net recommendations, with vertical dotted lines showing the diagnosis timepoint and key post therapy timepoints (3 and 12 months) in assessing optimal response to tyrosine kinase inhibitor treatment. Black dots represent the sampling points for whole genome sequencing.

Source Data

Extended Data Fig. 2 BCR::ABL1 event summary and exonic breakpoints exploration.

Panels a-c, show base pair resolution of the 3 fusions with exonic breakpoints: a. PD56961 b. PD57332 with an insertion of one nucleotide. Purple zigzag denotes the end of a fusion induced stop codon (TAA). c. PD57335 with an insertion of 6 nucleotides. d. Gel image of PD57335 BCR::ABL1 PCR product (well 5), showing that despite the breakpoint on chromosome 22 falling within exon 15, a standard e14a2 transcript was detected at the RNA level. Well 1 contains HyperLadder™ 100 bp (Meridian Bioscience) with positive control PCR products in wells 2–4 (E13A2, E14A2 and E1A2 respectively). The experiment was performed once. e. Splicing probabilities for PD57335 BCR::ABL1 fusion using SpliceAI. The vertical panels define 4 regions: the preceding exon to the chr22 breakpoint (BCR exon 14), the breakpoint within BCR exon 15, the intronic breakpoint within ABL1 intron1 and the following ABL1 exon 2. The rows describe: reference genes, reference splice probabilities between 0-1 for reference sequence with donor in green and acceptor in purple, fusion splice probabilities between 0-1 for reconstructed fusion sequence with donor in green and acceptor in purple, fusion sequence including 5 bp non-templated insertion, and fusion reading frame showing the consequence of the fusion sequence in grey, with the stop codon shown in red.

Source Data

Extended Data Fig. 3 Phylogenetic trees with branch-specific mutational signature profiles.

a. The graph shows the cosine similarity for single nucleotide variant (SNV) attribution to a combination of mutational signatures “SBS1”, “SBSblood” and “SBS18”, compared to attribution to SBS1 and SBSblood only. The change in the cosine similarity between the sigfit based posterior mean reconstructed signature and the observed signature is indicated by the directionality of the arrows, showing that for branches within the CML clade, particularly, early branches of the clonal expansion, as well as for early-in-life mutations, the cosine similarity improves with the inclusion of attribution to “SBS18”. Of note, there is little improvement in cosine similarity with the addition of SBS18 attribution for wildtype colonies or the trunk of the CML clade, ie., the pre-CML lineage, suggesting that SBS18 is not active outside of rapid expansion. Branches are categorised into five groups (i) branches of the BCR::ABL1-positive clonal expansion (“CML (mutant)”), (ii) ancestral branch of BCR::ABL1-colonies representing the lineage of origin of the CML (“Pre-CML (trunk)”), (iii) earliest mutations within the CML clade (ie., the shared branches within the clonal expansion, “Early CML (internal)”), (iv) branches of BCR::ABL1-negative colonies (“WildType, WT”), and (v) early-in-life branches (“Early life”). b. Phylogenetic trees of 9 patients with proportions of mutational signatures annotated to their branches. Bar plots show the proportion of SBS1, SBSblood and SBS18 contributing to SNVs spectra of different branch types. Bars represent posterior means with 95% credible intervals as error bars. The Y-axis shows the time in years since conception. The branches are categorised into four groups (i) branches of the BCR::ABL1-positive clonal expansion (“CML”), (ii) ancestral branch of BCR::ABL1-colonies representing the lineage of origin of the CML (“Pre CML”), (iii) the earliest mutations within the CML (ie., the shared branches within the clonal expansion, “Early CML”), and (iv) the branches of BCR::ABL1-negative colonies (wild-type, WT).

Source Data

Extended Data Fig. 4 Molecular time phylogenetic trees with branch-specific C > T at CpG proportions.

Phylogenetic trees of 9 patients with proportions of mutational signatures annotated to their branches. Bar plots show the proportion of C > T at CpGs contributing to SNVs of different branch types. The Y-axis shows the somatic mutation burden since conception. The branches are categorised into four groups (i) branches of the BCR::ABL1-positive clonal expansion (“CML”), (ii) the ancestral branch of BCR::ABL1-colonies representing the lineage of origin of the CML (“pre-CML”), (iii) the earliest mutations within the CML (ie., the shared branches within the clonal expansion, “early CML”), and (iv) the branches of BCR::ABL1-negative colonies (wild-type, WT).

Source Data

Extended Data Fig. 5 Validation of growth rates estimates.

a. Comparison of growth rates estimated using phylofit with two different maximum s values. b. Comparison of growth rates estimated using phylofit and the BirthDeathMCMC. Both methods are broadly consistent.

Source Data

Extended Data Fig. 6 BCR::ABL1 and CML in the All Of Us cohort.

a. CML diagnosis increases with BCR::ABL1 supporting read count. b. Each bar in the plot represents a BCR::ABL1 carrier identified, with the bar colour indicating the number of reads supporting their carrier status. Individuals diagnosed with CML or a pre-CML state during the follow-up period are marked with an asterisk. The Y-axis shows follow-up time, transformed using an arcsinh transformation (which approximates the natural logarithm while retaining zero-valued observations) for visualisation purposes. Exception received to the Data and Statistics Dissemination Policy from the All of Us Resource Access Board to display these data. c. CML incidence and BCR::ABL1 carrier status across age groups. In each panel, the bars represent the average number of new CML cases or BCR::ABL1 carriers per year, aligned with the left y-axis. The dots indicate the incidence rate per 100,000 population, corresponding to the right y-axis. For each age group, summary statistics were calculated by averaging annual statistics from 2018 to 2022, under the assumption that the population structure remained relatively stable during this period. WGS, whole genome sequencing subset of All of Us cohort.

Source Data

Supplementary information

Supplementary Information

Supplementary Notes 1–4.

Reporting Summary

Peer Review File

Supplementary Tables

Supplementary Tables 1–6.

Source data

Source Data Figs. 1–4 and Extended Data Figs. 1–6

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Kamizela, A.E., Leongamornlert, D., Williams, N. et al. Timing and trajectory of BCR::ABL1-driven chronic myeloid leukaemia. Nature 640, 982–990 (2025). https://doi.org/10.1038/s41586-025-08817-2

Download citation

Received: 25 February 2024
Accepted: 20 February 2025
Published: 09 April 2025
Issue date: 24 April 2025
DOI: https://doi.org/10.1038/s41586-025-08817-2

This article is cited by

Comparative assessment of the Sikun 2000 sequencing platform for whole genome sequencing
- Keya Cai
- Sibin Li
- Yongyong Ren
Scientific Reports (2025)

Subjects

Abstract

Similar content being viewed by others

Main

Driver mutations in CML colony genomes

Phylogenetic trees in patients with CML

Mutation accumulation in BCR::ABL1 cells

Telomere lengths in CML

Timing and rate of BCR::ABL1 expansion

BCR:ABL1 and CML in the All of Us cohort

Disease progression in CML

Discussion

Methods

Patients and sample acquisition

In vitro expansion of single-cell-derived blood colonies

RT–PCR for BCR::ABL1 transcript

DNA library preparation, sequencing and read alignment

Somatic mutation identification and filtering

Creating a genotype matrix

Phylogenetic tree topology

Donor with no wild-type colonies

Driver annotation

Timing branches

Quality control of a phylogenetic tree topology

Per sample BCR::ABL1 mutation status

Prediction of consequence of exonic BCR::ABL1 breakpoints

Mutational signature analysis

Growth rate estimation

Telomere analysis

Mixed models

SNV burden models

Telomere length models

Bulk DNA library preparation, sequencing and read alignment

Bulk unmatched somatic mutation identification and filtering

Bulk phylogeny reconstruction of PD60243

All of Us cohort analysis

Reporting summary

Data availability

Code availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Extended data figures and tables

Supplementary information

Source data

Rights and permissions

About this article

Cite this article

Share this article

This article is cited by

Search

Quick links