INDIRECT NUMBER NEEDED TO TREAT: COMPARATIVE EFFECTIVENESS OF GLASDEGIB + LOW-DOSE CYTARABINE VS AZACITIDINE OR DECITABINE IN ACUTE MYELOID LEUKEMIA PATIENTS INELIGIBLE FOR INTENSIVE CHEMOTHERAPY
Author(s): ,
Gabriel Tremblay
Affiliations:
Health Economics,Purple Squirrel Economics,New York,United States
,
Tracy Westley
Affiliations:
Health Economics,Purple Squirrel Economics,New York,United States
,
Shan Ashton Garib
Affiliations:
Health Economics,Purple Squirrel Economics,New York,United States
,
Timothy Bell
Affiliations:
Pfizer Inc,New York,United States
,
Joseph C Cappelleri
Affiliations:
Pfizer Inc,Groton,United States
,
Geoffrey Chan
Affiliations:
Pfizer Inc,Collegeville,United States
Anna Forsythe
Affiliations:
Purple Squirrel Economics,New York,United States
EHA Library. Forsythe A. Jun 15, 2019; 266662; PS1045
Dr. Anna Forsythe
Dr. Anna Forsythe
Contributions
Abstract

Abstract: PS1045

Type: Poster Presentation

Presentation during EHA24: On Saturday, June 15, 2019 from 17:30 - 19:00

Location: Poster area

Background
Recent U.S. approval of GLAS+LDAC improved treatment options for patients with acute myeloid leukemia (AML) for whom intensive chemotherapy is not an option due to older age or comorbidities. However, no clinical trials have directly tested glasdegib + low-dose cytarabine (GLAS+LDAC) against alternative AML therapies azacitidine (AZA) or decitabine (DEC). Number needed to treat (NNT) is a patient-centric measure familiar to clinicians: For overall survival (OS), NNT represents how many total patients need to receive one treatment over another in order to have one less death. Smaller NNT are desirable, with NNT=1 indicating the greatest relative treatment effectiveness. Previously, indirect methods in NNT compared landmark endpoints (time-to-event) between different trials in oncology, including proportion alive at 24 months (24-month OS) (Guyot, Cheng, Tremblay et al, 2018). 

Aims
In this study, indirect NNT applied 12-month OS to estimate how many patients treated with GLAS+LDAC instead of AZA or DEC would result in one less death.  

Methods
Phase II GLAS+LDAC vs LDAC patient data (n=116) were compared with separate published Phase III AZA vs LDAC (n=312) and DEC vs LDAC (n=485) OS results. Indirect NNT with 95% confidence intervals (CI) were derived from the absolute risk reduction (ARR) of death at 12-months (Casella and Berger, 2002). For reference, direct NNT were derived from within-trial ARR for each trial testing LDAC to GLAS+LDAC, AZA, or DEC (Sackett et al, 1996). 

Results

Within each trial, the direct NNT to prevent one patient death by 12-months was 3.23 (GLAS+LDAC vs LDAC), 6.90 (AZA vs LDAC) and 21.74 (DEC vs LDAC). With indirect NNT, for every three patients treated with GLAS+LDAC rather than AZA, one less would have died (NNT=-3.33). Likewise, for every two patients treated with GLAS+LDAC vs DEC, one less would have died (NNT=-2.45). Results with 95% CI conveyed statistical superiority of GLAS+LDAC vs AZA or DEC (Table 1).

Table 1. Within-trial and Indirect NNT Results for AML Treatments

Direct Trial Comparison 

Within-Trial 12-month OS 

Within-Trial ARR

Within-Trial NNT (95%CI) 

GLAS + LDAC vs LDAC

39.4% vs 8.4%

31.0%

3.23 (2.22, 5.87)

AZA vs LDAC

48.5% vs 34.0%

14.5%

6.90 (3.95, 27.10)

DEC vs LDAC*

32.3% vs 27.7%

4.6%

21.74 (-28.19, 7.84)

Intervention 

Indirect Comparator

Indirect ARR

Indirect NNT (95%CI)** 

GLAS + LDAC

AZA

30.0%

-3.33 (-14.14, -1.89)

GLAS + LDAC

DEC

40.8%

-2.45 (-6.56, -1.51)

*12-month OS for DEC vs LDAC was estimated from published Kaplan-Meier graphs (pooled 28 Supportive Care patients with 215 LDAC patients). **Negative values indicated benefit with GLAS+LDAC. In the text, absolute values were reported to aid interpretation.

Conclusion

Indirect NNT of 12-month OS demonstrated statistically significant findings for use of GLAS+LDAC over AZA or DEC. These results coincide with previously published Bucher indirect comparisons of GLAS+LDAC vs AZA or DEC (OS hazard ratios). With challenges in clinical trial representation of smaller populations, including older AML patients with comorbidities, indirect comparisons can be supportive of informative clinical decision-making and treatment prioritization.

Session topic: 4. Acute myeloid leukemia - Clinical

Keyword(s): Acute myeloid leukemia, Azacitidine, Cytarabine, Decitabine

By clicking “Accept Terms & all Cookies” or by continuing to browse, you agree to the storing of third-party cookies on your device to enhance your user experience and agree to the user terms and conditions of this learning management system (LMS).

Cookie Settings
Accept Terms & all Cookies