UCL Cancer Institute


Evolutionary trajectories of leukaemia development and therapy resistance in children

This project will build on our preliminary data to establish the potential co-symbiotic role of the bone marrow microenvironment in protecting resistant cells from treatment.

Applications are now closed for the 2019 postgraduate training programme.

  • Primary Supervisor: Prof Tariq Enver, Department of Cancer Biology, UCL
  • Secondary Supervisor: Dr Kathleen Curtius, Evolution and Cancer Laboratory, Barts Cancer Institute, QMUL

Funding note: Non-EU candidates are not eligible to apply 


Leukaemia is the most common childhood cancer and acute lymphoblastic leukaemia (ALL) accounts for about one-third of all childhood cancer deaths, almost on a par with brain tumours. Despite the overall good prognosis, standard chemotherapy treatment is arduous, and relapse in high-risk disease remains a major problem. Hence, dissecting the underlying cellular and molecular basis of tumour evolution to treatment resistance remains a priority. A better understanding of why childhood and adult ALL are different diseases with unique biological characteristics would also inform the ongoing clinical debate around whether adolescent and young adult (AYA) ALL patients should be treated as children or adults. Intratumor heterogeneity, both genetic and transcriptional/epigenetic, has been documented in many leukemias[1][2] and solid tumours, and the highly dynamic nature of tumour evolution likely reflects shifts in the selective pressures throughout leukaemia progression (namely: disease initiation, treatment response and relapse).
We have recently tested this hypothesis by exploring the relative contributions of genetic and epigenetic factors that characterize those leukaemic cells that persist in the immediate aftermath of induction chemotherapy. Our preliminary results suggest that, fuelled by parallel evolution and phenotypic convergence, a rare population of phenotypically uniform but genetically variegated cells escapes treatment, implying that phenotypic diversity rather than genetic diversity is the prime substrate for selection through induction chemotherapy in childhood-ALL5.

Project description

This project will build on our preliminary data to establish the potential co-symbiotic role of the bone marrow microenvironment in protecting resistant cells from treatment, and explore the similarity between treatment-resistant cells in AYA-ALL and childhood-ALL. The candidate will assess how the changing environment following treatment shapes clonal fitness, in particular whether genotype, phenotype, or a combination of both are key to selection during progression to relapse. Analysis of matching diagnostic and relapse samples shows only partial overlap between these both at the genomic and transcriptional level, suggesting that the relapsing disease might, at some point following treatment, transit through a stage of genetic homogeneity as a consequence of a clonal sweep [3].

The candidate will test this hypothesis by means of a xenograft model of transplantation and treatment previously established in the lab. The model will allow sequential sampling of the disease throughout progression, enable imaging of the resistant cells within the tumour microenvironment (TME), and afford the key distinction between deterministic and stochastic mechanisms of selection, since the same leukaemia can be transplanted into multiple recipients which can all be treated in the same way, and the post-treatment residuum in each compared.

A number of single cell sequencing and bioinformatics approaches will also be adopted to quantify heterogeneity at the level of genotypes and transcriptional networks. Candidate biomarkers of treatment-resistance and disease progression will then be validated in patient samples. In order to reconstruct the evolutionary trajectories toward resistance that best describe the serially collected genotype/phenotype data, the candidate will utilise mathematical modelling techniques such as Bayesian phylogenetics and stochastic branching processes to infer the timescales and rates governing convergent population dynamics. With these calibrated models, we can make statistical predictions about treatment response, and, based on the notion of “temporal collateral sensitivity”[4], inform hypotheses on how each distinct stage of tumour clonal evolution might be therapeutically exploitable.

Potential research placements

  1. Bioinformatics training in the analysis of single cell genomics data, supervised by Dr Peter Van Loo, The Francis Crick Institute.


  1. Anderson K. et al. Genetic variegation of clonal architecture and propagating cells in leukaemia. Nature, 201; 469, 356–361.
  2. Teschendorff AE, Enver T. Single-cell entropy for accurate estimation of differentiation potency from a cell’s transcriptome. Nat. Commun. 2017; Jun 1; 8: 15599.
  3. Mullighan CG et al. Genomic analysis of the clonal origins of relapsed acute lymphoblastic leukemia. Science, 2008; Nov 28; 322(5906): 1377-80.
  4. Zhao B et al. Exploiting Temporal Collateral Sensitivity in Tumor Clonal Evolution. Cell, 2016;  doi:10.1016/j.cell.2016.01.045