Venetoclax with Azacitidine or Low Dose Cytarabine As an Alternative to Intensive Chemotherapy in Fit Adults during the COVID19 Pandemic: Real World Data from the UK National Health Service

Background

Based on early evidence of a high rate of coronavirus mortality in patients with acute myeloid leukaemia (AML) undergoing intensive chemotherapy (IC), the national health service (NHS) in the United Kingdom temporarily made venetoclax available as an alternative therapy, with the aim of reducing both mortality and healthcare resource use. From late April 2020, venetoclax was available to patients aged >16y with NPM1 mutation without FLT3 internal tandem duplication (ITD), patients aged >50y with NPM1, IDH1 or IDH2 mutations (regardless of FLT3 status) and patients aged >60y without favourable-risk cytogenetics. Venetoclax could be given with either azacitidine or low-dose cytarabine (LDAC), with the latter recommended mainly for patients with NPM1 mutation. We report a health-system-wide real world data collection for toxicity and patient outcomes across 65 NHS Hospitals.

Methods

Each patient was registered on a central NHS database. Clinicians certified that their patient met the above criteria, had not received previous AML treatment, and was fit for induction chemotherapy. Anonymised data were retrospectively collected by treating physicians. Venetoclax dose, duration and toxicity information was requested for the first 4 cycles of therapy. Response definitions were as per European Leukaemia Network (ELN) guidelines. A total of 870 patients have been registered on the scheme, with outcomes reported here for those with follow-up information at a data cut on 1st August 2021.

Results

There were 301 patients, median age 72y (range 34 - 90) with 62% male. The majority (81%) had an ECOG performance status of 0-1. AML was secondary to a previous haematological disorder in 33%, therapy-related in 10% and de novo in the remaining 57%. MRC cytogenetic risk was intermediate in 70% and adverse in 27%. NPM1 mutations were detected in 28% and FLT3-ITD in 12%. Next-generation sequencing results were available in 86% of patients, which detected mutations in IDH1 or IDH2 in 28%, ASXL1 in 20%, RUNX1 in 17% and TP53 in 12%. The ELN risk was favourable for 23%, intermediate for 30% and adverse for 44%.

A majority received venetoclax in combination with azacitidine (85%), with the remaining 15% receiving LDAC. The LDAC cohort was enriched for de novo AML (76% vs 54%) and NPM1-mutated disease (56% vs 23%). Most patients (81%) followed the recommended initial schedule of venetoclax 100mg daily for 28 days in combination with posaconazole or voriconazole. Patients spent a median 14 days in hospital in cycle 1, then a median of 0 days for cycles 2-4. In cycles 1, 2, 3 and 4, the median number of days for recovery of neutrophils to >0.5x10 ⁹/L was 33, 25, 24 and 14 respectively, and the median number of days to recovery of platelets to >50x10 ⁹/L was 22, 3, 0 (no drop below 50) and 0.

The composite complete remission (CR) / CR with incomplete haematological recovery (CRi) rate was 70%. MRD data is being collected. The best response was morphological leukaemia free state (MLFS) in 2%, partial remission in 7% and refractory disease in 11%. CR/CRi was higher in de novo (78%) compared to secondary AML (57%, p=0.02); NPM1 mutated (78% vs 67%, p=0.02) and IDH1/IDH2 mutated disease (85% vs 62%, p=0.02). ELN favourable risk patients had the highest CR/CRi rate (85%, intermediate 71%, adverse 60%, p=0.01).

Median follow-up was 8.2 months (95%CI 7.8 - 9.0) with median overall survival (OS) 12.8 months (95%CI 10.9 - not reached). Mortality at day 30 was 5.7% and day 60 was 8.4%. 12-month overall survival was 51%, increasing to 71% in those who achieved CR/CRi. Survival was poorer in secondary (HR 1.9, p <0.01) and therapy-related AML (HR 2.1, p=0.02), better in NPM1 mutated (HR 0.6, p=0.02) and IDH mutated (HR 0.5, p=0.02) disease and poorer with TP53 mutation (HR 2.0, p=0.01). Overall survival did not differ for patients treated with LDAC compared to azacitidine (HR 1.1, p=0.7).

Conclusion

This large real-world study demonstrates CR/CRi and survival rates comparable to those reported in prospective clinical trials. Importantly, during the COVID-19 pandemic, the adoption of venetoclax regimens permitted the great majority of treatment to be delivered as an outpatient with significant resource saving at a time of critically constrained inpatient resources. The data support prospective comparisons of venetoclax-based regimens to IC in fit adults with AML particularly in older patients with de novo AML, NPM1-mutated and IDH-mutated disease.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal