Literature Scan
Sorafenib Plus Standard Chemotherapy Associated
with Improved Event-Free Survival in Acute Myeloid
Leukemia, But Also Toxicity
36
ASH Clinical News
CI 4-15). The three-year EFS rate was
similarly higher among sorafenib-treated patients: 40 percent (95% CI 29-51)
versus 22 percent (95% CI 13-32), for a
hazard ratio (HR) of 0.65 (95% CI 0.450.91; p=0.013; FIGURE).
The benefits of sorafenib were
sustained even after adjusting for established prognostic factors (including lactate dehydrogenase concentration, white
blood cell count, and secondary acute
myeloid leukemia), with an adjusted
HR for EFS of 0.64 (95% CI 0.41-0.99;
p=0.0434). “A favorable cytogenetic
profile and NPM1 mutation were associated with better event-free survival,” the
authors wrote, “whereas FLT3 internal
tandem duplication status remained a
negative prognostic factor.”
In an explorative analysis, Dr. Röllig
and colleagues analyzed the 46 patients
with FLT3 mutations, finding no clear
pattern for median EFS (5 months [95%
CI 0-12] in the sorafenib group and 6
months [95% CI 1-11] in the placebo
group), these patients did experience
longer relapse-free survival and overall
survival with sorafenib:
relapse-free survival in the sorafenib
group remained significant.”
The rates of complete remission
(CR) were similar between both groups:
81 (60%) patients in the sorafenib
group and 78 (59%) patients in the
placebo group. Median relapse-free survival among patients who achieved CR
was 32 months, with 47 percent (95%
CI 39-56) of patients remaining relapsefree at three years (56% in the sorafenib
group, 38% in the placebo group).
The most common treatment-related grade 3 or 4 adverse events included
fever, infections, pneumonia, and pain,
with grade ≥3 treatment-related adverse events occurring more frequently
in the sorafenib cohort.
Fifty-eight adverse events led to
withdrawal from study treatment (40 in
the sorafenib group and 18 in the placebo
group). Adverse events most commonly
associated with study withdrawal were:
• Infections (13 [31%] with
sorafenib vs. 7 [37%] with placebo)
• Skin toxicity (10 [24%] vs. 2
[11%])
• Hematologic toxicity (4 [10%] vs.
2 [11%])
• Gastrointestinal toxicity (5 [12%]
vs. 1 [5%])
• Cardiac events (2 [5%] vs. 2
[11%])
• Pain (4 [10%] vs. 0)
• Bleeding (1 [2%] vs. 2 [11%])
• Liver toxicity (1 [2%] vs. 1 [5%])
• Hypertension (0 vs. 1 [5%])
• Relapse-free survival: 18 months
(95% CI 0-36) with sorafenib
versus 6 months (95% CI 0-16)
with placebo
• Overall survival: not reached with
sorafenib versus 19 months (95%
CI 0-39) with placebo
However, these differences were not
statistically significant. “If patients
with FLT3 internal tandem duplication mutations were excluded from the
study population,” the authors noted,
“the increases in event-free survival and
After accounting for prognostic factors,
the HR for relapse or death favored
sorafenib over placebo at 0.48 (95% CI
0.27-0.85; p=0.011). And, after three
years of follow-up, 63 percent (95%
CI 54-72) of patients in the sorafenib
group and 56 percent (95% CI 47-65)
of patients in placebo group were alive.
Three treatment-related deaths were reported in the placebo group (two cases
of pneumonia and one case of toxic
liver failure) and four were reported in
the sorafenib group (two cases of pneumonia and two other infections).
No patients died within two weeks
of the first induction cycle, with similar
30-day and 60-day mortality rates
between the sora fenib and placebo
groups: 2 percent versus 1 percent and
4 percent versus 4 percent, respectively.
“Since early mortality, the proportion of patients who achieved complete
remission, and the number of deaths in
complete remission were similar in both
treatment groups, the longer eventfree survival in the sorafenib group is
mainly attributable to a reduction in
relapses and an increase in the time in
complete remission,” the researchers
noted. “As there was no improvement in
overall survival in the sorafenib group
compared with the placebo group, it is
unclear whether salvage treatment was
equally potent in patients relapsing after
placebo or sorafenib treatment.”
“Longer follow-up of this trial and
results of other ongoing studies are
needed to establish whether the benefit
in relapse-free survival translates into
improved overall survival,” Dr. Röllig
told ASH Clinical News. ●
REFERENCE
Röllig C, Serve H, Huttmann A, et al. Addition of sorafenib versus
placebo to standard therapy in patients aged 60 years or younger with
newly diagnosed acute myeloid leukaemia (SORAML): a multicentre,
phase 2, randomised controlled trial. Lancet Oncol. 2015 November 5.
[Epub ahead of print]
Event-Free Survival (A), Relapse-Free Survival (B), and Overall Survival (C)
A
100
Sorafenib
Placebo
Relapse-free survival (%)
80
60
40
Censored:
HR=0·64 (95% CI 0·45–0·91); p=0·012
Not censored:
HR=0·66 (95% CI 0·48–0·90); p=0·01
20
0
0
Number at risk
Sorafenib 134
Placebo 133
12
24
36
Time since randomisation (months)
72
61
60
46
38
25
B
C
100
100
80
80
Overall survival (%)
FIGURE.
Event-free survival (%)
Adding the multi-kinase inhibitor
sorafenib to standard chemotherapy
increased event-free survival (EFS)
among patients with newly diagnosed
acute myeloid leukemia (AML), according to results of a recent randomized, double-blind, placebo-controlled,
phase II study conducted by Christoph
Röllig, MD, of the Dresden University of Technology in Germany, and
colleagues. However, its use was also
associated with increased toxicity.
“When this trial began in 2009,
standard treatment for AML consisted
of a combination of cytarabine plus
anthracyclin/anthracenedione. The
need for improvement was obvious,
with only around 50 percent of patients
surviving – even among younger
patients,” Dr. Röllig told ASH Clinical
News. “To our knowledge, this is the
first randomized, controlled trial showing that integrating a kinase inhibitor
into standard intensive chemotherapy
of younger patients with AML is associated with significant improvement of
relapse-free survival, with no increase
in treatment-related mortality.”
The trial included 267 patients (age
range = 18-60 years) with newly diagnosed de novo or secondary acute AML
who were enrolled at 25 sites across
Germany between March 27, 2009, and
November 28, 2011.
Forty-six (17%) of the patients were
positive for FLT3 mutations and 86
(32%) had NPM1 mutations.
Patients were randomized 1:1 to receive standard chemotherapy plus either
sorafenib 400 mg twice daily (n=134)
or
Articles
placebo (n=133). Standard chemotherapy
consisted of two cycles of induction
therapy (daunorubicin 60 mg/m2 on days
3-5 plus cytarabine 100 mg/m2 on days
1-7), followed by three cycles of postremission therapy (high-dose cytarabine
3 mg/m2 twice daily on days 1, 3, and 5).
“Prognostic factors such as age, white
blood cell count, and lactate dehydrogenase concentration at initial diagnosis, de
novo AML, and cytogenetic and molecular risk factors were similar between both
groups,” the authors noted.
After a median follow-up of 36
months, 137 events (primary treatment
failure, relapse, or death) had occurred,
with an overall median EFS of 13.5
months (95% CI 9-18). Treatment with
sorafenib led to greater median EFS than
placebo: 21 months in the sorafenib
group (95% CI 9-32) compared with
nine months in the placebo group (95%
60
40
Censored:
HR=0·53 (95% CI 0·30–0·93); p=0·025
Not censored:
HR=0·57 (95% CI 0·36–0·89); p=0·017
20
0
0
58
53
48
36
40
Censored:
HR=1·04 (95% CI 0·65–1·66); p=0·884
Not censored:
HR=0·86 (95% CI 0·58–1·27); p=0·382
20
12
24
36
Time since randomisation (months)
81
78
60
14
26
0
0
134
133
12
24
36
Time since randomisation (months)
91
100
77
80
51
51
Figure 3: Event-free survival (A), relapse-free survival (B), and overall survival (C)
Data censored and not censored for allogeneic stem-cell transplantation. HR=hazard ratio.
p=0·465). To explore the reason for the smaller difference
between groups in overall survival compared with
relapse-free survival, we did another post-hoc subgroup
analysis of overall survival after relapse, which showed
followed by skin toxicity (ten [24%] vs two [11%]),
January 2016
haematotoxicity (four [10%] vs two [11%]), gastrointestinal
toxicity (five [12%] vs one [5%]), cardiac events (two [5%]
vs two [11%]), pain (four [10%] vs none), bleeding (one