The study I’m going to present is based on a small subset of patients treated on a clinical trial called UKALL2011, all of whom had an ABL-class fusion. To start with, I’d like to remind everyone that many years ago now a group of patients were identified by gene expression profiling that had a gene expression profile similar to BCR-ABL but lacked that canonical gene fusion. A number of years later a network of genes called the ABL-class fusions were identified. These all mimic the functional consequence of BCR-ABL but obviously they don’t have that fusion themselves. So they have fusions of ABL1, ABL2, CSF1R and PDGFR-β or –α. An ABL-class fusion involves one of these hub genes as a [pre-prime partner gene and a variable five prime partner gene]. So it’s a true network of abnormalities but the key thing is that they all have constitutive activation of ABL. As a result, many people have postulated that they are sensitive to treatment with a tyrosine kinase inhibitor such as imatinib or dasatinib.
Indeed, there is considerable in vitro and in vivo evidence to support this. There is also emerging clinical evidence that patients with one of these fusions will benefit from imatinib but because of their rarity there are very few, if any, studies that have managed to directly compare the outcome of patients with an ABL-class fusion who receive imatinib and have not received imatinib. This is what we sought to do.
We had noticed, like many others, that patients with an ABL-class fusion had a very high risk of relapse and also very high levels of residual disease at the end of induction. So on our trial, UKALL2011, we prospectively screened patients who had MRD levels at the end of induction of more than 1%, so this is a very selected cohort. Where they were positive we then supplemented their therapy with imatinib.
Because this screening programme started after the start of the 2011 trial, at the end with a combination of retrospective screening to supplement the prospective screening, we now have two groups of patients all of which have an ABL-class fusion, one which received imatinib in first remission and one who did not. So overall the frequency of patients with an ABL-class fusion in this selected cohort of patients that had MRD of more than 1% at the end of induction was 11-13% in T- and B-cell precursor ALL. So this is across the spectrum.
We identified a total of 21 patients on this trial but they had a variety of different ABL-class fusions. 10 of them, i.e. just under 50%, had EBF1/PDGFR-β and that is consistent with what other people, other studies, have shown, that this is by far the most prevalent single fusion.
So, in terms of treatment groups, as I have alluded to, there were two treatment groups. There was the early TKI group that comprised 13 patients. In this group of patients the ABL-class fusion was always detected in and around the end of induction and all patients received imatinib very soon after the ABL-class fusion was identified. This was usually within two weeks of the end of induction, so fairly early on. As I said, they all received imatinib and they all received additional chemotherapy as well. Most of these patients, three-quarters, went on to receive a transplant as well.
The control group was smaller, comprised of eight patients. None of these patients received imatinib in first remission and all of them, like the other cohort, received additional chemotherapy. So the two groups were comparable in terms of their sex ratio, their median age, their white cell count and also their TCR status and roughly the distribution of EBF1/PDGFR-β. Obviously all these patients had high levels of MRD at the end of induction but for the patients with these ABL-class fusions 1% was the very low end. The average was well over 20%, so the vast majority of these patients had strong evidence of resistant disease. By the current criteria, primary refractory disease of more than 5%, three-quarters of them would have been classified as having primary refractory disease.
When we compared the outcome of the group of patients that received early TKI with the others there was significant difference. The patients that received a TKI early in their therapy there were no relapses, that’s zero out of 13 patients had a relapse after a median follow-up time of about 2½ years. Importantly, two of the patients did die, sadly, after transplant complications. In contrast, of the eight patients that did not receive a TKI in first remission, six out of these eight patients either relapsed or died of primary refractory disease. So this was a clear-cut difference in outcome.
Interestingly, three of the patients that relapsed in this control cohort did receive a TKI in second remission after their relapse but sadly two of the three then went on to die, indicating possibly that imatinib at relapse is not as effective as imatinib in initial diagnosis.
So, overall this study provides additional evidence that if you give patients with an ABL-class fusion a tyrosine kinase inhibitor alongside high dose chemotherapy, so none of these patients had imatinib as a sole agent, it does improve outcome. There are a number of caveats, of course, with this study – the numbers are small but this is a rare subgroup of patients. The follow-up time is modest and, of course, all patients received additional consolidation therapy of varying types. However, on a positive note, there were very few reported cases of grade 3 or grade 4 toxicity and very few of them, there was only one case, where it was possibly attributable to the imatinib. So we believe that giving imatinib to most of these patients is safe.
We’re taking this research forward and in our new trial, which will be the ALLTogether-01 trial which will recruit patients across Europe from 14 different countries, we will screen all patients for ABL-class fusions at diagnosis and add imatinib to their therapy at day 15 for those that are under the age of 60. The idea behind this is we want to reduce the high levels of MRD that have previously been seen in previous cohorts by giving the imatinib from day 15. So we will see with time whether or not this does provide not only further improvement but it does avert these high levels of MRD at the end of induction.