Highlights from

EHA 2019

European Hematology Association

Amsterdam 13-16 June 2019

Overcoming the “Don’t Eat Me” signal in AML and MDS

Antileukaemic activity was observed with anti-CD47 macrophage checkpoint inhibitor 5F9, both as monotherapy and in combination azacytidine, in patients with acute myeloid leukaemia (AML) and myelodysplastic syndrome (MDS), as presented by Dr David Sallman, (Moffitt Cancer Center, USA) [1].

5F9 is a novel form of immunotherapy that targets CD47 to restore innate immunity, the first line of defence against cancer cells. CD47 provides the notorious “don’t eat me” signal displayed by many types of cancer cells that enables macrophage immune evasion by preventing phagocytosis. CD47 is the dominant macrophage checkpoint that is overexpressed in most cancer cells and increased CD47 expression has been associated with poorer prognosis.

Therefore, the 5F9005 study compared the activity of 5F9 administered alone or in combination with azacitidine in patients with AML or MDS. In addition to patients with relapsed/refractory AML, the study enrolled patients with previously untreated AML who were ineligible for induction chemotherapy and untreated MDS patients at intermediate to very high risk by IPSS-R criteria. Ten patients with relapsed/refractory AML received sole 5F9 at 30 mg/kg twice weekly and 36 treatment-naive patients were treated with the standard regimen of azacitidine (75 mg/m2 on days 1 to 7) plus a priming dose of 5F9 of 1 mg/kg followed by ramp-up to 30 mg/kg given weekly. The priming dose was used to mitigate on-target anaemia.

In the cohort of patients with relapsed/refractory disease, 6 patients with AML and 4 with MDS received 5F9 monotherapy. These patients had received a median of 2 (range 1 to 6) prior therapies. In the cohort of patients with untreated AML or MDS, 5F9 was administered with azacitidine as first line to 22 patients with AML and 17 patients with MDS. In the respective combination therapy arms, the median patient age was 74 and 71 years. Fifty-five percent of patients with AML also had underlying myelodysplasia. Among MDS patients, 41% of patients were at high risk and 12% of patients were at very high risk.

The primary objectives of the study were to evaluate the safety and efficacy of 5F9 treatment in relapsed/refractory AML and 5F9 in combination with azacitidine in patients with untreated AML or MDS. The safety analysis showed 5F9 was well tolerated both as monotherapy (n=10) and in combination (n=36). The maximum tolerated dose was not reached in either treatment arm, and the safety profile of 5F9 combined with azacitidine was similar to that of single-agent azacitidine. In the combination arm, 1 dose-limiting toxicity, grade 4 hemagglutination, was observed that resolved within 24 hours. One patient discontinued combination treatment due to an adverse event. No cytopenia, infections, or autoimmune adverse events were observed, and no deaths occurred within the initial 60 days of treatment.

Antileukaemic activity was observed with 5F9 monotherapy and in combination in patients with AML and MDS. Among the 10 patients in the relapsed/refractory AML cohort receiving 5F9 monotherapy, the objective response rate was 10%, which represented a marrow complete response (CR); 70% of patients demonstrated stable disease and 20% experienced disease progression.

In the cohort of 14 patients with AML and 11 patients with MDS receiving 5F9 plus azacitidine as first line therapy, the ORR was 64% and 100%, respectively. In the AML subgroup, the responses included 5 (36%) patients with CR, 2 (14%) patients with morphologic complete remission with incomplete blood count recovery, and 2 (14%) patients with marrow CR. Stable response was seen in 36% of patients, and no patients had disease progression. Of the 11 responding patients with MDS, 6 (55%) patients had CR, 4 (36%) had marrow CR, and 1 (9%) patient showed haematologic improvement. Again, no disease progression occurred. The time to response was more rapid at 1.9 months with the combination than that observed with azacitidine alone.

The median duration of response was not reached (range 0.03+ to 8.3+) months. In patients with MDS treated with the combination, 43% had a complete cytogenetic response, and 20% showed MRD negativity. Median duration of response was not reached (range 0.5+ to 4.3+). The median follow-up in both groups was short, approximately 3.8 months, and is ongoing. “The longest patient in response is in CR for 9+ months in therapy and ongoing,” Dr Sallman noted. “Combination treatment also eliminated leukemic mutations in some AML patients and eliminated leukaemia stem cells in responding AML and MDS patients.”

  1. Sallman D, et al. The first-in-class anti-CD47 antibody HU5F9-G4 is active and well tolerated alone or in combination with azacitidine in AML and MDS patients: initial phase Ib results. Abstract S878, 24th 24th Congress of the European Hematology Association, 13-16 June 2019, Amsterdam, the Netherlands.

The content and interpretation of these conference highlights are the views and comments of the speakers/authors.