Disappointing negative trial results have dashed hopes for the thrombopopoietin receptor agonist eltrombopag as adjuvant therapy in patients with acute myeloid leukaemia.
Combining eltrombopag with standard induction therapy of daunorubicin and cytarabine appeared to do more harm than good in AML patients, according to results from a phase 2 randomised double blind trial conducted in 10 international cancer centres including Victoria and NSW.
Despite promising results seen with eltrombopag in previous phase 1-2 trials of patients with myelodysplastic syndrome and AML, the latest RCT published in Lancet Haematology showed that eltro use showed a trend for more serious adverse events and deaths due to haemorrhage, as well as shorter survival, compared to placebo.
In the study, 148 treatment-naïve patients were randomised to receive eltrombopag 200mg or placebo starting on day 4 of standard induction therapy (daunorubicin plus cytarabine). Treatment was continued until platelet counts were 200×109/L or higher, until remission or after 42 days from the start of chemotherapy.
For the primary endpoint, serious adverse events were reported in 24 (32%) of patients in the eltrombopag group and in 14 (20%) of the placebo group. It was notable that more grade-3-5 bleeding events occurred in the eltrombopag group (9% vs 1%) and there were five early deaths due to haemorrhage in the eltrombopag group and none in the placebo group.
The study authors also pointed to the higher numbers of deaths in the eltrombopag group attributable to AML (26% vs 14%), deaths within 30 days of the last dose of study drug (15% vs 6%).
Overall, 39 (53%) of patients in the eltrombopag group died versus 29 (41%) in the place group.
For the secondary endpoints of haematological recovery and survival, patients in the eltrombopag group also fared no better and sometimes did worse than the placebo group. Blood count recovery was no faster and platelet transfusion outcomes were similar. At two years follow up, the durations of response to eltrombopag were shorter and media overall survival was 15.4 months vs 25.7 months for placebo.
The study investigators – who included Professor Jeffery Szer of the Royal Melbourne Hospital, said the reasons for the negative results with eltrombopag were no clear, particularly since earlier studies had suggested that it may have an additional antileukaemic effect by inhibiting haematological malignant growth.
An accompanying commentary said the results highlighted the importance of doing sufficiently powered comparative trials of therapies that seem to have a rational case for efficacy.
It noted that results are awaited from another phase 2 trial of eltrombopag in combination with decitabine in older AML patients not eligible for induction chemotjerapy.
Until then, “the use of eltrombopag in combination with induction chemotherapy, as administered in the present study, cannot yet be endorsed for acute myeloid leukaemia,’ it concluded.