After how many years from diagnosis/completing treatment should children and young adults with acute lymphoblastic leukaemia be considered "cured"? Clinical trials generally report 5 year event free survival because the risk of relapse after that time-point is very low and this is assumed to mean cure. Indeed, the risk of relapse after four years from completing therapy was <1% for patients treated on St Jude's trial Total Therapy XV and they proposed that length of follow-up should be considered to be the "time to cure". The group had previously shown that excess deaths beyond ten years in non-irradiated patients was similar to the general population whereas irradiated patients had a greater risk of excess mortality due to the development of secondary neoplasms. However, there was limited information about the time to cure and relative survival in sub-groups with variable relapse risk. We examined the long-term outcome of patients treated on an MRD stratified protocol UKALL 2003 that recruited patients between 2003-2011 to determine time to cure for the overall trial population and prognostic sub-groups defined by age, sex, immunophenotype, cytogenetics and MRD response.

There were a total of 3113 patients eligible for analysis with a median follow up of 9.4 years (only 20 patients received cranial irradiation). Relative survival rates were estimated using nationwide population mortality rates as the baseline accounting for age, sex and calendar year (Office of National Statistics 2018 release). The relative survival of UKALL2003 patients compared to equivalent children in the UK population was 96% (95% CI 95.5-96.7). Relative survival did not differ by sex but was significantly worse for patients >10 years compared with younger patients (p<0.001). As expected, patients with good risk cytogenetics ( ETV6-RUNX1, and high hyperdiploidy) and negative bone marrow Ig/TCR MRD at end of induction (EOI) had better relative survival rates and lower excess mortality (all p<0.001) compared with patients who had adverse risk factors (high risk genetics and EOI MRD+).

To estimate the time to a cure, we calculated the conditional probability of relapse at 12 years for patients who had survived event to the start of each follow-up year. Although the initial risk of relapse differ significantly by sex, age, MRD and genetics the risk of relapse for all subgroups quickly coalesced at around 5-6 years (Figure 1). Furthermore, the time to cure, defined as a relapse risk <1% (dashed blue line of figure), was similar across all the subgroups at 6-7 years.

In conclusion, the relative survival of young patients with good risk cytogenetics and EOI MRD negative status approaches that of their normal peers. However, regardless of prognosis, the time to cure is similar across risk groups.

Figure 1
Figure 1.
View largeDownload slide

Disclosures

No relevant conflicts of interest to declare.

© 2021 by The American Society of Hematology
2021
Sign in via your Institution