Report on Scientific Research into Acute lymphoblastic leukaemia (ALL), supported by CCT’s Stopcancer.health programme. 9 12.2019

Professor Ian N Hampson, Dr Xiaotong He and Dr Lynne Hampson,

University of Manchester Viral Oncology Labs., Research Floor, St Mary’s Hospital, Oxford Road,

Manchester M13 9WL

Acute lymphoblastic leukaemia (ALL) is the most common children’s cancer with the majority of cases occurring between 3 and 4 years of age and it has been suspected for >100 years that an unidentified infectious agent may cause this disease [1-4]. The main difficulty is that it has not been possible to identify either current or past infections in patients without first knowing what to test for. Comparing antibody repertoires might provide a solution to this problem although this is extremely challenging as a single human produces ≈10 billion different antibodies as cumulative historical record of exposure to past and current infectious agents. Advantages of this approach would include:- (a) High sensitivity since the immune response amplifies any reaction to disease; (b) Identification of silent infections which produce no symptoms; (c) The identity of so called ‘Hit and Run’ infections which cause damage which persists after they are killed by the immune system; (d) The ability to detect human proteins which are wrongly targeted by anti-self (autoimmune) autoantibodies.

We now report development and use of the Serum Antibody Repertoire Analysis (SARA) technique to compare antibodies found in pooled blood samples from 96 children with ALL with those found in pooled blood from 100 normal control (NC) children. The output identifies infections in addition to any human proteins that stimulated an immune response in either ALL or NC children. Statistical adjustment of the results was carried out to give the highest scoring unique organisms in each pool which showed there were 44 in ALL and 40 in NC children. Analysis of the type of organisms detected in both groups showed that overall there were nearly 4 times as many viruses detected in ALL rather than NC children with the majority of these belonging to members of the class known as single-stranded RNA (ssRNA) viruses (eg Rabies, JEV etc). Further analysis of one of these normally only found in bovines, confirmed the presence of antibodies to this virus, or a close relative, in a high proportion of individual ALL blood samples. There was also a marked change in antibodies to the type of bacteria present with ALL children having reduced numbers of the beneficial types of probiotic bacteria and increased numbers of disease associated types when compared to those found in the NC children. With respect to autoimmune targeted human proteins, >1000 were detected in ALL children and yet remarkably none were found in the NC group. It is also highly significant that many of the identified ALL antibody-targeted proteins have known functions relevant to leukaemia such as control of cell growth and cell death and the development of blood cells, [5-7].

In summary, SARA has provided strong evidence for the existence of a hitherto unidentified ssRNA virus as a causative agent for ALL. Furthermore, the results also support a protective role for specific groups of bacteria combined with a detrimental role for other species. Most unexpected was the finding of autoantibodies against human proteins with documented roles in controlling the production of blood cells, exclusively in children with ALL. Indeed this is the first evidence for an autoimmune component to this disease. Overall our results provide valuable new insights on the causes of childhood ALL and suggest new approaches to both its prevention and treatment.

For the future, we are working with a group based in Mexico City which has one of the highest incidences of ALL in the world. They are providing us with blood samples obtained from children with ALL before any treatment whereas the blood samples used previously were obtained after treatment. Analysis of these samples with SARA should allow us to discriminate between differences which are related to either treatment of ALL or the disease itself. This should identify factors involved in both causation of ALL and/or its response to treatment.

References

1. McNally, R.J.Q. and T.O.B. Eden, An infectious aetiology for childhood acute leukaemia: a review of the evidence. British Journal of Haematology, 2004. 127(3): p. 243-263.
2. Dockerty, J.D., Epidemiology of childhood leukemia in New Zealand: Studies of infectious hypotheses. Blood Cells Molecules and Diseases, 2009. 42(2): p. 113-116.
3. Eden, T., Aetiology of childhood leukaemia. Cancer Treatment Reviews, 2010. 36(4): p. 286-297.
4. Greaves, M., A causal mechanism for childhood acute lymphoblastic leukaemia. Nature Reviews Cancer, 2018. 18(8): p. 471-484.
5. Futterer, A., et al., Dido gene expression alterations are implicated in the induction of hematological myeloid neoplasms. Journal of Clinical Investigation, 2005. 115(9): p. 2351-2362.
6. Guo, C.C., et al., KMT2D maintains neoplastic cell proliferation and global histone H3 lysine 4 monomethylation. Oncotarget, 2013. 4(11): p. 2144-2153.
7. Li, Y.H., et al., Role of a non-canonical splice variant of the Helios gene in the differentiation of acute lymphoblastic leukemic T cells. Oncology Letters, 2018. 15(5): p. 6957-6966.