Prague, Czech Republic: In "Dr Jekyll and Mr Hyde" Robert Louis Stevenson wrote about the good and evil sides of the same person; now scientists in Australia have discovered that in acute lymphoblastic leukaemia (ALL) there are Dr Jekyll and Mr Hyde cells - "good" and "evil" clones of the same type of ALL cell.

The "evil" cells are clones that have a pre-existing, rather than acquired, resistance to drugs used for treating ALL, and their presence in a patient means that person will inevitably relapse after chemotherapy, however well they respond initially.

The discovery means that now researchers may be able to design therapies that will specifically target these resistant subclones so that, in the future, patients who have been identified as having them can be treated immediately with the alternative therapies.

ALL is the most common cancer in children and, although nearly all patients will respond initially to chemotherapy, one in four will relapse. Seoyeon Choi told the EORTC-NCI-AACR [1] Symposium on Molecular Targets and Cancer Therapeutics in Prague today (Thursday 9 November): "We have previously shown that these relapses were due to small numbers of surviving and highly drug refractory cells. However, until now, it has been unclear whether these relapses resulted from the acquisition of therapy-induced drug resistance or were caused by a subpopulation of cells that were already intrinsically drug resistant."

Ms Choi, a final year PhD student at the Children's Cancer Institute Australia in Sydney and medical student at the University of Sydney, Australia, analysed samples taken from 25 ALL patients at the time of their diagnosis and at their relapse to discover the molecular "fingerprint" of every ALL cell.

"White blood cells, or lymphocytes, are unique in that every one has its own molecular signature. Therefore, we can 'molecular fingerprint' each lymphocyte in order to know what the leukaemia 'looks' like. We found 'fingerprints', or clonal markers, that revealed the emergence or evolution of new clonal populations at the time of relapse in 13 patients. In eight of the samples, highly sensitive clone-specific PCR [polymerase chain reaction] revealed that these 'relapse' clones had been present in small numbers at the time of diagnosis, indicating that they were involved in the mechanism of relapse.

"My research indicates that these are not different leukaemias, but a smaller population of related cells that are naturally more aggressive than the major clone. The problem is that they are present at such low levels, hidden behind the obvious leukaemia; the patient would appear to be responding well to treatment with the major leukaemia clones dying, but, in fact, the small number of subclones can survive therapy and cause a relapse."

The researchers found that the presence of the subclone at diagnosis correlated significantly with the length of the first clinical remission, and the more of the subclone that was present, the quicker the patient relapsed.

Ms Choi said: "I believe it is important to know that these cells are actually more resistant and aggressive from the very beginning, like a evil twin, if you like. While the 'good' twin, or the major clone, appears to be responding well - and lulling the clinicians into a false sense of security - the 'evil' twin, or subclone that is identified too late because of their small numbers, can cause relapse, by which time they are present in very high numbers. Knowing this, we can identify such patients early on in their treatment and focus on new therapies that target the right cell/clonal population so that we may be able to improve the outcome for this special group of patients who relapse early.

"Patients who relapse early usually have a particularly poor outcome, and if we could prevent the relapse that is inevitable under the current treatment regime, then we might be able to make a big difference to these children's survival."

The researchers believe that, in those patients where they were unable to detect subclones, the cells may have existed, but in numbers too low to be detected by current methods, and they are working to improve the PCR technique in order to increase its sensitivity so that it can be used to identify even smaller numbers of subclones.

At present there are no therapies that can be used specifically to treat children who are identified as having the relapse subclone. However, Ms Choi said that now her research had identified the cells that made relapse inevitable, it would be possible for researchers to start work on therapies that could target these cells. "If we could treat these differently, by targeting them early in therapy, or introducing alternative therapies, we may improve the overall outcome of every patient with leukaemia.

"While I do not know when this research will translate into clinical changes, I do believe that it will happen in my lifetime when I graduate from medical school and start to practice as a physician."

Abstract no: 252

1. EORTC [European Organisation for Research and Treatment of Cancer, NCI [National Cancer Institute], AACR [American Association for Cancer Research].

EORTC-NCI-AACR