©TheCanadian Journal ofUrology™: International Supplement, April 2014
compared to13.6months (95%CI, 11.3months to15.8
months) in the placebo cohort. This improvement
inmedian survival by 4.8months, corresponding to
a 37% reduction in the risk of death comparedwith
placebo,wasdeterminedwhen thestudywasstopped
early at a planned interim analysis (HR for death in
enzalutamide group 0.63, p < 0.001). Enzalutamide
treatment led tosuperioroutcomes inprostate-specific
antigen (PSA)reduction>50%, radiographicresponse
rate, time to progression and time to first skeletal
related event over placebo.
Toxicityratesbetween the twogroupsweresimilar,
despite a significantly longer on treatment time for
men in the enzalutamide cohort. More men in the
enzalutamide arm experienced fatigue, diarrhea,
hot flashes, musculoskeletal pain, headache and
hypertension. Ofnote,fivepatientsintheenzalutamide
cohort experienced seizure activity, with possible
predisposing comorbid brain metastasis, organic
brain disease, and adverse drug interaction cited as
possible contributing factors. In this and in ongoing
trials,patientswithhistoryofbrainmetastasis, seizure,
head trauma with loss of consciousness, transient
ischemic attack in the last 12 months, stroke, brain
arteriovenous malformation, or use of concomitant
medicationswhich could lower the seizure threshold
wereexcluded, and thus thesafetyofenzalutamide in
thesepopulations isnot known, seeTable 1.
Which subsets of patients benefit from
enzalutamide?
The cohorts in theAFFIRM studywerewellmatched
for all factors at baseline, including by Gleason
grade, with median Gleason grade of 8 in each
group, andGleason grade > 7 in 50.4% and 52.4% in
the enzalutamide and placebo cohorts respectively.
The benefit of enzalutamidewas seen across all pre-
specifiedsubgroups, includingthose<65versus65and
older, bygeographic treatment location, baselinepain
score and type of disease progression at study entry
(PSA or radiographic). Post-hoc subgroup analyses
demonstratedsimilarbenefitofenzalutamide inmen<
75versus75andolder, aswell asbenefit in thosewith
liverand lungmetastasiswhencompared toplacebo.
4,5
Clinical benefit, assessed by health related quality of
life scores,was significantlybetter formen treatedon
enzalutamide,with improvements inphysical, social,
emotionalandfunctionalwell-beingcomparedtothose
treatedwithplacebo.
6
Evaluationofpatientswhowere
found tobe long term responders, on studyagent for
>12or>18months,werenoted tohave less baseline
disease burden, longer time from cancer diagnosis to
study enrollment, and improved ratesof biochemical
andradiographicresponse toenzalutamidecompared
to thoseon study<12months.
7
Multivariateanalysis
of hazard ratio for death demonstrated survival
advantage for those with ECOG performance status
0 or 1 compared to 2, lower baseline pain score, PSA
as compared to radiographicprogression, novisceral
disease, lower values of LDH and higher values of
hemoglobinatstudyentry.
3
Gleasongradeatdiagnosis
was not included in thismultivariate analysis due to
substantial missing data, thus the effect of Gleason
grade upon efficacy of enzalutamide post docetaxel
isnot known.
Should steroidsbeprescribedconcomitantly
with enzalutamide?
Many men treated post docetaxel are on long term
steroid therapy, andmay represent a fundamentally
different population thanmen not on, or who have
not progressed on steroids. The authors sought to
understanddifferencesbetweenpatientswithdisease
progressionon steroidsat enrollment (approximately
30% ineach cohort), compared to thosewhowerenot
on steroids upon outcomes in theAFFIRM study in
post-hoc analyses.
8
Amultivariate analysis showed
medianoverallsurvivalwas11monthsversusmedian
survival not met inmenwith baseline corticosteroid
usecompared to thosenotonbaselinesteroids,despite
study treatment group. By study group, patients in
the enzalutamide cohort on corticosteroids had a
median overall survival of 12.3months compared to
9.3months on placebo, and this difference remained
statistically significant.
Following trial enrollment,mennot on steroids at
baselinewere alsopermitted to initiate corticosteroid
therapyat investigatordiscretion,and thus theeffectof
allonstudyuseofcorticosteroidswasalsoevaluated.
9
The combined baseline and on study initiation of
steroidswas 48% in the ezalutamide and 45% in the
placebo group. The median survival in all patients
treatedwithonstudycorticosteroidswas11.5months,
andnotmet inthosenotoncorticosteroids. Statistically
significantbenefitof enzalutamideoverplacebo inall
outcomemeasures was retained despite steroid use.
Notably,grade3and4adverseeventrateswerehigher
in all patients on corticosteroids. Though baseline
prognostic factorswere reported to be slightly better
inpatientsnotoncorticosteroids, theauthorscontend
that steroidusemay be associatedwithunmeasured
or unidentified disease factors or other properties of
steroid use. Thesemay include promotion of tumor
growthvia aberrantmutantAR activation.
10
65
Hoffman-CensitsandKelly
