General AML

EHA 2019 | Results from phase III ASTRAL-1 study of guadecitabine versus treatment of choice 

On Saturday 15 June 2019, during the 24th meeting of the European Hematology Association (EHA) in Amsterdam, NL, the phase III results of the ASTRAL-1 study were presented by Professor Pierre Fenaux, Hôpital Saint Louis, Paris, FR.1

The ASTRAL-1 study compared the use of guadecitabine (SGI-110) to treatment of choice (ToC) in patients with acute myeloid leukemia (AML) who had not received prior treatment and who were not eligible for intensive chemotherapy.

Background and phase II results
  • Guadecitabine is a next generation hypomethylating agent 
    • Dinucleotide of decitabine results in slow deamination by cytidine deaminase, longer half-life and prolonged exposure to the active metabolite, decitabine

A previous phase II trial (NCT01261312) enrolled 103 patients; the total responses of these patients, by dose and length of administration are shown in Table 1.2

Table 1: Response rates to guadecitabine by dose and length of administration in phase II trial

Length of administration

5 days

10 days

-

Dose of G

60mg/m2

90mg/m2

60mg/m2

Total

N

24

27

52

103

Complete response (CR)

38%

41%

33%

36%

CR with incomplete neutrophil recovery regardless of platelets (CRi)

17%

11%

8%

10%

CR with incomplete platelet recovery (CRp)

0%

7%

10%

7%

Composite CR (CRc)

54%

59%

50%

53%

Based on the results in Table 1, investigators decided to use a 60mg/m2 dose of guadecitabine, administered over 5 days, for the phase III trial.

Phase III study design and patient characteristics
  • Randomized 1:1 to either G or ToC
    • G: 60mg/m2 subcutaneously for 5 days
    • ToC: decitabine (DEC), azacitidine (AZA), or low dose Ara-C (LDAC)
    • Table 2 shows treatments given pre- and post-randomization
  • Primary endpoints: CR rate and overall survival (OS) rate
  • The majority of patients were unfit and 75 years or older as shown in Table 3
  • In total, 815 patients were randomized, with 793 treated (largest trial in treatment naïve patients with AML)

Table 2: Treatment given pre- and post-randomization

Randomization

DEC

G

AZA

G

LDAC

G

Pre-

351 (43%)

-

340 (42%)

-

124 (15%)

-

Post-

173

178

178

162

56

68

Table 3: Patient characteristics

 

G

(n = 408)

ToC

(n = 407)

Median age

76 (56–93)

76 (59–94)

≥75 years old

62.0%

62.4%

ECOG status 2 or 3

50.5%

50.3%

Secondary AML

36.3%

36.9%

Poor-risk cytogenetics

34.3%

34.6%

FLT3-internal tandem duplication (ITD)

7.1%

7.8%

NPM1 mutation

16.2%

14.5%

CEBPA (biallelic) mutation

1.0%

3.2%

TP53 mutation

12.5%

10.5%

Bone marrow blasts >30%

71.3%

68.1%

White blood cells ≥20,000μl

15.2%

14.3%

Efficacy

Primary endpoint analysis in the intention-to-treat (ITT) population is shown in Table 4 (CR and OS for guadecitabine versus ToC).

Table 4: Primary endpoint analysis in ITT population

 

G

ToC

P value

CR

79

(19.4%)

71

(17.4%)

0.48

Median time to CR (months)

4.5

(1.9–19.1)

4.4

(1.9–15.1)

-

Median OS (months)

7.10

8.47

0.73

HR (95% CI): 0.97 (0.83–1.14)

12-month OS

37%

36%

-

24-month OS

18%

14%

-

In the subgroup analysis, there were no differences between treatment arms based on age, sex, cytogenetic risk, ECOG score, or race. The exception was in patients with TP53 mutations who appeared to favor the control arm (ToC).

The survival analysis (Table 5) shows improved outcome for patients responding to ≥4 cycles or ≥6 cycles guadecitabine compared to ToC. This superior survival was more pronounced when patients received >6 cycles (P = 0.002) compared to >4 cycles (P = 0.02) of guadecitabine. These results indicate that prolonged exposure to guadecitabine is necessary to observe clinical benefit.

Table 5: Survival of patients with a response (CR, CRp, CRi or PR) who received ≥4 cycles and ≥6 cycles of guadecitabine

 

G

ToC

P value

More than 4 cycles (n = 476)

0.02

HR (95% CI): 0.78 (0.64–0.96)

Median survival

15.6 months

13.0 months

12-month survival

60%

52%

24-month survival

29%

20%

More than 6 cycles (n = 375)

 

Median survival

19.5 months

14.9 months

0.002

HR (95% CI): 0.69 (0.54–0.88)

12-month survival

75%

62%

24-month survival

37%

24%

Adherence to treatment and safety

Treatment exposure is shown in Table 6, with reasons for discontinuation of treatment in patients receiving less than 4 cycles shown in Table 7 with the safety data on AEs and serious AEs (SAEs) shown in Table 8. As shown in Table 6, approximately 41% and 54% of patients did not receive 4, or 6 cycles, respectively.

Table 6: Treatment exposure

 

G (n = 401)

ToC (n = 392)

Median number of cycles received

5.0

5.0

Patients receiving <4 cycles

42.4%

40.8%

Patients receiving < 6 cycles

54.2%

53.8%

Table 7: Reasons for discontinuation of treatment (patients receiving <4 cycles)

  G
ToC G ToC

Number of cycles

<4 

<4 

<6 

<6 

Death

17.6%

15.7%

22.5%

18.9%

Progressive disease

7.6%

7.6%

10.8%

12.0%

Adverse event (AE)

6.4%

5.2%

7.4%

6.4%

Patient decision

5.7%

5.4%

6.6%

7.6%

Randomized, not treated

1.7%

3.7%

1.7%

3.7%

Alternative anti-leukemia therapy pursued

0.7%

0.5%

0.7%

1.2%

Lost to follow-up

0.2%

0.0%

0.5%

0.0%

Other

2.5%

2.7%

3.9%

3.9%

Total

42.4%

40.8%

54.2%

53.8%

Table 8: Safety data of AEs

 

G (n = 401)

ToC (n = 392)

AE leading to discontinuation

41 (10.2%)

26 (6.6%)

SAE

325 (81.0%)

296 (75.5%)

Death due to AE

115 (28.7%)

117 (29.8%)

Grade ≥3 AE

367 (91.5%)

343 (87.5%)

The most common grade 3 and 4 AEs occurring in ≥20% of patients in either group were: febrile neutropenia, pneumonia, thrombocytopenia, neutropenia and anemia.

Conclusion

This study failed to meet the primary endpoint of a statistical difference in CR and OS between guadecitabine and ToC. However, a benefit was observed in patients who received >4 cycles, indicating that treatment duration is crucial to response. Around 40% of patient in both arms did not receive the minimum of 4 cycles to see the maximum of treatment benefit.

In relation to safety, both guadecitabine and all three ToC options were comparable, though higher rates of febrile neutropenia and pneumonia were noted in the guadecitabine arm.

References
  1. Fenaux P. et al. Results of ASTRAL-1, a phase 3 randomized trial of guadecitabine vs treatment choice (TC) in treatment-naïve acute myeloid leukemia not eligible for intensive chemotherapy. Oral abstract #S879. 2019 Jun 15. European Hematology Association (EHA) 24th Annual Meeting, Amsterdam, NL.
  2. Kantarjian H.M. et al.Guadecitabine (SGI-110) in treatment-naive patients with acute myeloid leukaemia: phase 2 results from a multicentre, randomised, phase 1/2 trial. Lancet Onc. 2017 Aug 24. DOI: 10.1016/S1470-2045(17)30576-4
Download this article:

You can now download this article in Adobe PDF® format.

Download as PDF
Was this article informative? Thank you for your feedback!