Ravulizumab provides long-term control of intravascular haemolysis and improves survival in patients with PNH
Background: Increasing genomic stratification coupledwith an evolving therapeutic landscape present challenges inthe choice of optimal induction therapy in adults with acutemyeloid leukaemia (AML). We have pioneered an innovativealgorithmic approach, combining drug eligibility criteriaand genetic risk stratification, which facilitates the develop-ment of consensus recommendations for front-line intensiveAML therapy in the UK (Coats et al, BJH 2021). In light ofthe publication of the revised ELN 2022 AML guidelines(Dohner et al, Blood 2022) we have repeated these analyses.
Methodology: 1000 in silico AML cases were generated tocover the spectrum of clinical and genetic features, but up-dated to include information on FLT3-TKD and TP53 mu-tation status. The eligibility criteria for each NHS-fundedintensive treatment and the ELN2022 (including an addi-tional TP53mut and complex karyotype [CK] adverse riskcategory) were converted into a digital format. These criteriaand classifications were used to assign the in silico cases todistinct clinical scenarios based on the combination of ELNrisk group and the choice(s) of available funded treatments.A representative case from each scenario was independentlyreviewed by 11 members of the UK AML Working Groupseeking their preferred induction treatment in a fit 40- anda 65-year-old patient over two rounds of Delphi consensus.For consensus, 75% of respondents needed to agree.Frequency of each clinical scenario, as a percentage of allcases, was estimated from UK AML trial data.
Results: 1000 cases were assigned to 24 different clinicalscenarios using our digital algorithm. This compared to 22scenarios in our previous consensus guidance and reflectedchanges in the ELN 2022 which included changes in the prog-nostic impact of FLT3-ITD, FLT3-TKD and TP53mut + CKadverse risk group. Delphi consensus was undertaken forthe 24 scenarios to ascertain whether clinical practice wouldchange in line with the new classification.For the patient aged 40, a consensus was reached for 15/24scenarios representing 96.7% of cases. For the patient aged65, a consensus was reached for 15/24 scenarios represent-ing 95.8% of cases. The recommendations from the previousconsensus changed in 5.2% of cases.
Conclusions: We have confirmed that our methodology iseffective for generating consensus UK treatment recommen-dations reflecting changes in the ELN 2022 treatment rec-ommendations. This is available as a webapp https://amlconsensus.er.kcl.ac.uk. The presented methodology has thepotential to generate consensus treatment guidelines for pa-tients with other blood cancers and solid tumours.